1 Interconnectio within Hong Kong · Web viewGovernment of the Hong Kong SAR Interconnection and Competition in the Hong Kong Electricity Supply Sector October 1999 Reference C1789

FINAL REPORT

Government of the Hong Kong SAR

Interconnection and Competition in the Hong Kong Electricity Supply Sector

October 1999

Reference C1789

This report has been prepared by Environmental Resources Management the trading name of Environmental Resources Management Limited, with all reasonable skill, care and diligence within the terms of the Contract with the client, incorporating our General Terms and Conditions of Business and taking account of the resources devoted to it by agreement with the client.

We disclaim any responsibility to the client and others in respect of any matters outside the scope of the above.

This report is confidential to the client and we accept no responsibility of whatsoever nature to third parties to whom this report, or any part thereof, is made known. Any such party relies on the report at their own risk.

CONTENTS

EXECUTIVE SUMMARY i

1 OVERVIEW i2 INTERCONNECTION iv3 COMPETITION ix4 SOUTHERN CHINA ELECTRICITY SECTOR xi5 CONCLUSIONS AND RECOMMENDATIONS xii

FOREWORD xvGLOSSARY xviABBREVIATIONS xxi

1 INTRODUCTION AND OVERVIEW 1

1.1 BACKGROUND 11.2 STUDY OBJECTIVE 31.3 STUDY APPROACH 31.4 STRUCTURE OF THIS DOCUMENT 101.5 KEY TERMS 11

2 FEASIBILITY OF INCREASED INTERCONNECTION OF ELECTRICITY SYSTEMS IN HONG KONG 13

2.1 INTRODUCTION 132.2 GENERAL CONSIDERATIONS ASSOCIATED WITH ELECTRICITY

INTERCONNECTION 142.3 METHODOLOGY AND APPROACH 182.4 TECHNICAL FEASIBILITY OF INCREASED INTERCONNECTION 262.5 ECONOMIC FEASIBILITY OF INCREASED HEC-CLP

INTERCONNECTION 472.6 CONSUMER BILL AND TARIFF IMPACTS OF INCREASED HEC-CLP

INTERCONNECTION 552.7 KEY CONCLUSIONS ON INTERCONNECTION 62

3 FEASIBILITY OF THE INTRODUCTION OF ELECTRICITY SECTOR COMPETITION IN HONG KONG 65

3.1 INTRODUCTION 65

ERM GOVERNMENT OF THE HKSAR

3.2 LOCAL AND WIDER CONTEXT FOR ELECTRICITY SECTOR COMPETITION 66

3.3 EVALUATION OF FOUR MARKET STRUCTURES APPLIED TO HONG KONG 89

3.4 DETAILED CONSIDERATION OF OPTIONS APPROPRIATE TO HONG KONG 97

3.5 KEY CONCLUSIONS ON COMPETITION 111

4 DEVELOPMENTS IN THE SOUTHERN CHINA ELECTRICITY SECTOR 113

4.1 INTRODUCTION 1134.2 GUANGDONG ELECTRICITY SUPPLY INDUSTRY 1134.3 PROCESS OF RESTRUCTURING 1164.4 IMPLICATIONS FOR HONG KONG 1184.5 KEY CONCLUSIONS ON CO-ORDINATION WITH SOUTHERN CHINA 123

5 CONCLUSIONS AND RECOMMENDATIONS 125

5.1 INTRODUCTION 1255.2 CONCLUSIONS ON THE TECHNICAL FEASIBILITY OF INCREASED

INTERCONNECTION 1265.3 CONCLUSIONS ON ECONOMIC ANALYSIS OF

INTERCONNECTION/GENERATION SCENARIOS 1275.4 CONCLUSIONS ON LOGISTICAL CONSIDERATIONS ASSOCIATED

WITH IMPLEMENTING INCREASED INTERCONNECTION 1325.5 WIDER IMPLICATIONS OF THE INTERCONNECTION/GENERATION

SCENARIOS 1335.6 CONCLUSIONS ON FINANCIAL AND TARIFF ANALYSIS OF

INTERCONNECTION CASES 1355.7 CONCLUSIONS ON ANALYSIS OF FUTURE COMPETITION OPTIONS 1375.8 RECOMMENDATIONS 139

6 CASE STUDIES 143

CASE STUDY 1: PROJECTING COMPETITIVE MARKET PRICING IN THE USA145

CASE STUDY 2: INTERNATIONAL EXPERIENCE OF ELECTRICITY COMPETITION — USA, UK, BRAZIL AND THE EU 149

CASE STUDY 3: HOW MANY PLAYERS ARE NEEDED FOR A MARKET TO BE GENUINELY COMPETITIVE? 155

CASE STUDY 4: STRANDED COSTS163


CASE STUDY 5: COMPETITION IN THE TELECOMMUNICATIONS INDUSTRY IN HONG KONG167

CASE STUDY 6: PRICE SHOCKS IN THE US WHOLESALE MARKET173

LIST OF TABLES

Table 1.3.a Accuracy of Data and Information in Each Period of the Study Time-frame 7

Table 1.5.a Key Terms Used in this Study 11Table 2.2.a Spectrum of Benefits available from Considering Firm

Capacity of Interconnection 16Table 2.2.b Spectrum of Benefits Available from Considering Firm Power

Transfer Across Interconnection 17Table 2.4.a Interconnector Decoupling Incidents in Hong Kong 27Table 2.4.b Reduction Achieved in Spinning Reserve Due to Existing

Interconnector 37Table 2.4.c Historical Fuel Cost Savings from Economy Exchanges 41Table 2.4.d Summary of Additional Benefits Potentially Available from

Increased Interconnection Capacity 42Table 2.5.a Discounted Present Values of Cumulative Incremental

Capital Expenditures for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP 50

Table 2.6.a Discounted Present Values of Cumulative SCA Revenue Requirements for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP 58

Table 3.2.a Background Situation in Hong Kong and Mainland China 68Table 3.2.b Advantages and Disadvantages of Each Market Structure 79Table 3.2.c Comparison of Key Characteristics between the Electricity

Sector in Hong Kong and Countries that have Introduced Forms of Competition 82

Table 3.3.a Potential Limitations on Competition in Generation in Hong Kong 91

Table 3.4.a Market Transition Trajectories in Hong Kong 107Table 3.4.b Windows of Opportunity for Implementation 110Table 4.2.a Comparison of Guangdong and Hong Kong Electricity Sectors

(1997) 114Table 4.2.b Capacity in the Three Provinces of South-West China 115Table 4.4.a Implications for Hong Kong of a Restructured Guangdong

Electricity Sector 120


LIST OF FIGURES

Figure 1.1.a The Hong Kong and South China Interconnected System Configuration 2

Figure 2.2.a Potential Overall Benefits of Interconnected Systems 17Figure 2.5.a Discounted Present Values of Cumulative Incremental

Capital Expenditures for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP 49

Figure 2.6.a Annual Differences in SCA Revenue Requirements Between Each Interconnection and Generation Planning Scenario and the Base Case Consolidated Across HEC and CLP 57

Figure 2.6.b Annual Tariff Differences Between Each Interconnection and Generation Planning Scenario and the Base Case Consolidated Across HEC and CLP 59

Figure 2.6.c Annual Tariff Differences Between Interconnection/Generation Scenario B and the Base Case Consolidated Across HEC and CLP and for HEC and CLP Separately 60

Figure 3.2.a Market Structure 1 — Monopoly / the Vertically Integrated Utility 75

Figure 3.2.b Market Structure 2 — Single Wholesale Purchaser 76Figure 3.2.c Market Structure 3 — Multiple Wholesale Purchasers 76Figure 3.2.d Market Structure 4 — Retail Competition 77Figure 3.4.a Market Structure 1 in Hong Kong — Two Vertically Integrated

Monopolies 105Figure 3.4.b Market Structure 2 in Hong Kong — Single Wholesale

Purchaser 106Figure 3.4.c Market Structure 3 in Hong Kong — Multiple Wholesale Purchasers

106Figure 4.3.a Planned Process of Restructuring for Guangdong Electricity Sector

117

LIST OF BOXES

Box 3.2.a Northern Ireland’s Experience of Electricity Restructuring 87Box 3.2.b Scotland’s Experience of Electricity Restructuring 88Box 3.2.c Ukraine’s Experience of Electricity Restructuring 89


NOTE

A Chinese translation of the Executive Summary of this report is available as a separate document.


FOREWORD This report presents the results of a feasibility study of whether consumer benefits might be gained in Hong Kong’s electricity sector from increases in interconnection capacity between The Hongkong Electric Company (HEC) and CLP Hong Kong Limited (CLP) and the introduction of competition. Quantitative economic analysis of potential cost reductions from increased interconnection, expressed in terms of both total consumer benefits and the effect on tariffs was conducted at a level of detail appropriate for a feasibility study. Qualitative assessment was made of various electricity sector structures, including competitive market models. The assessment of the potential economic benefits of increased interconnection required the use of actual utility data, of available projections of peak load growth and of plant capital, fuel and operating costs. The use of such actual data specific to Hong Kong’s electricity sector improves the usefulness of this Study (compared with some alternative approaches such as generic international comparisons alone). However, several points must be borne in mind. This Study adopted a long-term (30-year) time horizon in order to focus on the medium- to long-term benefits potentially available. The utilities’ detailed generation and transmission expansion planning horizons are considerably shorter than this (in the order of 10 to 15 years, which is standard utility practice in most countries). The focus of this Study is therefore on long term policy questions, not on short-term utility planning.


xv

GLOSSARY

Term Description

Available Transfer Capability (ATC)

As used in this Study: the amount of interconnection capacity available on the interconnector once the largest circuit outage contingency has been allowed for (N-1) and capacity has been set aside to allow inter-area spinning reserve obligations to be met.

Bill saving The reduction in electricity bills (summed across all customers) associated with a reduction in revenue requirement. In this Study, such reductions arise in the interconnection/generation scenarios due to reductions in capital expenditure requirements and associated reductions in absolute profit levels under the Scheme of Control framework.

Bus or bus-bar An electrical conductor to which are attached a number or circuits.

Captive customer

A customer who has no choice of supplier.

Carrying charges

See transmission fees.

Consolidated In this Study, the combined effect of economic and/or financial results on HEC and CLP and on the customers of the two companies. Key results are also presented for the two companies separately.

Contingencies A contingency is an event that could cause a disruption in the supply of electricity. Examples include the loss of output from a power station or loss of power flow across an interconnector circuit, which may occur due to protection equipment built into the system being “tripped.” Allowance in system planning for any single contingency is called an N-1 criteria, where there are N relevant elements (such as generators or transmission circuits) in the system. Allowance for double (two simultaneous) contingencies is known as an N-2 criteria.

Distribution The process of transforming electricity from high voltage to low voltage and distributing that energy to consumers.

Diversity in load profiles

Load diversity is a measure of difference between load profiles, where a load profile describes the variation in load or electrical demand over a period of time, such as a day or a year. If two given load profiles have high diversity, then they are dissimilar such that they have non-coincident peaks — one tends to have high load values at the time that the other has low values and vice versa — and when added together the resulting peak demand will be considerably smaller than the sum of the individual peak demands. In most electricity systems benefits of diversity exist between customer groups — many commercial customer loads peak during the day while residential customer loads peak during the evening. When the load profiles in two adjoining areas are diverse — as may happen when one that has a predominantly commercial load peaking in the middle of the day and another that has a predominantly residential load that peaks at night — opportunities exist for synergy, because the generation capacity necessary to meet the peak of the interconnected areas as a whole is lower than that required to meet each area individually. There is not a great deal of diversity between the HEC and CLP load profiles, so the opportunity for such synergy in Hong Kong


xvi

Term Descriptionis limited.

Firm Capacity As used in this Study: the technical capability of the interconnection itself to provide “firm” inter-area generation capacity support as and when it is needed. (Note: this term is sometimes used to refer to generation capacity contracted between a supplier and a customer on a firm basis such that it could be considered the same as if the customer had ownership over the capacity. The existing Interconnector Agreement in Hong Kong uses a much more limited definition of firm capacity than this.)

Firm Power As used in this Study: the technical capability of the interconnection itself to support the regular, reliable transfer of energy between areas, as would be required if generation plant was jointly dispatched for the system as a whole. (Note: this term is sometimes used to refer to contractual arrangements between utilities for the supply of energy. The use of the term in this Study does not necessarily entail such arrangements, although neither does it preclude such arrangements being put in place.)

Generation The process of generating electricity. Also “generation plant:” power station units; “generation capacity:” maximum available instantaneous output from a power plant unit or collection of units.

Hot standby Generation capacity that must be able to be online within 30 minutes should spinning reserve not be able to meet a contingency.

Independent Power Producer (IPP)

IPPs are generally thought of as electricity generators who are not utility affiliated, eg this will not apply to Black Point or Lamma power stations under the current regulatory structure, but potentially a new power station in Hong Kong or elsewhere. Utilities may become IPPs if there is separation of ownership of generation from the transmission and distribution of electricity.

Independent System Administrator (ISA)

A variant on the Independent System Operator model for transmission system operation.

Independent System Operator (ISO)

Potentially separate from the owner of the transmission system, an ISO is the manager of the transmission system and often determines dispatch order, undertakes planning functions and ensures fair and non-discriminatory access to the transmission system. Generally considered a requirement for a competitive market.

Independent Transmission Company (ITC)

A variant on the Independent System Operation model for transmission system operation.

Independent Transmission System Operator

Same as Independent System Operator.

Interconnection Essentially a transmission line, but between two utilities.Joint Dispatch Centre (JDC)

Physical facility with control equipment for dispatching simultaneously the generation plants of more than one utility.

Joint dispatch Generally: dispatch of power plants that is undertaken in co-ordination with other utilities. As used in this Study: Operating or “dispatching” generation plants in separate areas on a joint basis to realise any available synergy and operational


xvii

Term Descriptioneconomies to achieve the least cost generation production cost.

Joint planning As used in this Study: generation and transmission planning that is undertaken jointly between neighbouring utilities, taking into account interconnection options in order to achieve least-cost expansion of the system as a whole.

Load centres Centres of high load density, eg Central, Hong Kong.Load shedding The process of cutting the supply of electricity to customers.LOLE Loss of load expectation is the inverse of LOLP.LOLP Loss of load probability, or the likelihood that generation

capacity will not be sufficient to supply electricity to all customers who demand it.

Loose pool Referred to as Joint Planning in this Study, whereby two areas are interconnected to form one wider system and capacity additions are planned jointly taking into account the interconnection to allow load to be served at lower cost than would be possible for the systems independently, given a specified level of reliability.

Outage The planned withdrawal from service (usually for maintenance) or forced unavailability (unplanned interruption due to faults) of a generation unit, or any part of a transmission or distribution system, for a period of time. Usually measured in terms of weeks per year for planned maintenance outages and fraction of time or probability of unavailability for forced outages.

Performance-Based Regulation (PBR)

In a typical PBR system, prices are linked to a key economic index adjusted for expected increased in productivity. Generally used as an alternative to Rate of Return type of regulation.

Power marketer Within a market that has implemented retail competition, it is unlikely that small consumers will be willing to undertake the complex transactions required to arrange supply from a generator. Consequently, power marketers have emerged that will perform this function for small consumers.

Pooling arrangements

The set of arrangements or rules under which a particular power pool operates. Pooling arrangements can be constituted on a simple co-operative basis, as has been done for decades in some parts of the United States, or it can form the core of a wholesale competitive market. In the first case, it involves simple resource-sharing as defined by the agreements for the operation of the pool. In the second case, generation companies bid power into the pool, an independent power exchange or system operator with joint dispatch facilities dispatches the plant in economic merit order and wholesale purchasers (distribution companies who also retail energy, independent retail energy brokers and large customers) purchase energy from the pool at the real time ‘spot price’. In such competitive markets, the wholesale buyers and sellers usually also “hedge” against movements in the spot price via contracts. This Study deals in some detail with simple pooling arrangements such as Firm Capacity and Firm Power. Issues associated with more sophisticated pool arrangements and operations appropriate for a full competitive wholesale market will not be of concern for many years and so are not dealt with in detail in this Study.

Power pool When two or more utilities act together co-operatively or under the auspices of an independent entity to supply electricity through either joint planning of new facilities or joint


xviii

Term Descriptiondispatching of their power stations.

Rate-of-Return (RoR) Regulation

A type of regulation that allows utilities to earn a rate of return on net fixed assets, as the Schemes of Control Agreement in Hong Kong allow.

Reliability The ability of the system to meet customer demand for energy. Full reliability analysis involves consideration of all three levels in the supply chain: Level I generation, Level II transmission and Level III distribution.

Reserve margins The sum of all available generating capacity, including firm capacity commitments, divided by the utility’s maximum demand, minus one (-1). The reserve margin is one of the key indicators of whether a utility has sufficient generating capacity in cases of emergency.

Retail competition

One step beyond Third Party Access whereby all customers have access to the transmission system.

Revenue requirement

The revenue required by a utility from the total of customer bill payments to cover all costs and allow a given level of profit.

Short, medium and long term

For the purposes of this Study, the periods are defined as 5 to 10, 10 to 20 and more than 20 years respectively.

Single buyer A type of electricity market structure whereby one organisation has a monopoly over the wholesale purchase and retail sale of all electricity. Usually considered a transition step towards a competitive market.

Stranded investment/costs

Usually defined as prudently incurred costs, usually approved by governments, that cannot be recovered in a changed regulatory environment or a competitive market. Such costs can be significant in issue when changes are made to electricity systems, because plant and infrastructure for electricity supply have relatively long service lives and involve investments which are recovered over a relatively long time frame. Where the physical assets associated with a stranded investment are held by an outside company, the stranded costs may appear in the form of a contract or contractual commitment within an electricity utility.

Supply/Retailing The process of organising the physical supply of electricity from generation through transmission and distribution to final consumers. This is usually only discussed as a distinct activity in the context of a competitive market. Especially in a competitive market, this process may not necessarily be performed by the companies involved in the generation, transmission or distribution of electricity, but instead by an independent third party.

System security The ability of a system to continue to meet demand, following an abnormal occurrence, without overloading any component part of that system.

System stability The stability of a power system is characterised principally by the maintenance of frequency, voltage and power transfer within limits. It is also dependent on the design and operation of the transmission system, the response of generation units to disturbances and the availability of reactive power.

Take-or-pay A contractual obligation which requires a company to pay for some specified amount of a commodity (such as electricity or fuel), whether or not the company actually takes that quantity. Such contractual obligations may be simple or complex. It is possible for such contractual arrangements to allow some


xix

Term Descriptiondegree of flexibility with respect to the timing of consumption of the commodity, although not usually with respect to payment for it.

Tariff adjustment/ effect/impact

The average effect on tariffs of a change in revenue requirement, calculated as the change in revenue requirement in dollars (positive or negative) divided by energy consumption in kiloWatt-hours.

Third Party Access

A type of electricity market structure whereby third-party generation companies — other than the company that owns the network — are granted access to that network to supply electricity to customers. The option for customers to obtain supply from third party generation companies (via bi-lateral contracts) is usually limited by size. Without Third Party Access arrangements, a generation company has a virtual monopoly over the customers to whom they have connected the electricity network, as it is usually either not economic or not practical to construct a parallel network to supply customers in competition with an incumbent generation/transmission/distribution company.

Tie or tie-line See interconnector.Tight pool Referred to as Joint Planning with Joint Dispatch in this Study.

The addition to a loose pool arrangement whereby the dispatch of all units included in the multi-area system is undertaken in economic order (subject to certain specified conditions or constraints).

Total Transfer Capability (TTC)

As used in this Study: the amount of interconnection capacity available on the interconnector once the largest circuit outage contingency has been allowed for (N-1).

Transformer A device used to step electricity from one voltage level to another. Transmission and distribution networks progressively step power down in standard increments from high voltage levels (which allow low current and therefore low loss transmission) to lower voltage levels (which ultimately allows distribution to customers at the standard supply voltage level required by their equipment).

Transmission fees

The fees charged by the owner of a transmission system to carry electricity from generators to final customers.

Transmission The process of transmitting electricity, at high voltage, from generators to distributors.

Unbundling Refers to the separation of functions previously undertaken by a single entity, and/or the separation of prices for various aspects of electricity supply. For example, the functions of an ISO and/or Single Buyer are required to be operationally separate from those of generation and distribution. Steps towards unbundling include the accounting separation of vertically integrated utilities.

Vertical integration

Full vertical integration involves the combination of the functions of generation, transmission, distribution and energy retailing within a single electricity utility company. The two separate electricity companies in Hong Kong are fully vertically integrated.

Wheeling of power

Another term for the transfer of power across transmission lines, usually by third parties who must pay fees for the use of the transmission system.


xx


xxi

ABBREVIATIONSAbbreviation

Full Text

AC Alternating currentAPEC Asia Pacific Economic Co-operationAVR Automatic Voltage Regulation CLP CLP Power Ltd (formerly China Light & Power Company Ltd)CPI-X Consumer Price Index minus X, a method of price regulationDC Direct currentEEI Edison Electric Institute, an association of electric utilities in the USAEHV Extra high voltageERM Environmental Resources ManagementFERC Federal Energy Regulatory CommissionFOR Forced Outage RateGPHC Guangdong Electric Power Holding Company in Mainland ChinaGWh GigaWatt-hourHEC Hongkong Electric Company LtdIPP Independent Power ProducerISA Independent System Administrator ISO Independent System OperatorITC Independent Transmission Company JDC Joint Dispatch Centre kV KiloVoltkW KiloWattkWh KiloWatt-hourLOLE Loss-of-load-expectationLOLP Loss-of-load probabilityLRMC Long-Run Marginal Costs MVA MegaVolt-AmpsMW MegaWattMWh MegaWatt-hourNERC North American Electricity Reliability Council, a USA based organised

of utilities which focuses on electricity reliability issuesNTS Nuclear Transmission SystemPAT Profits after taxesPBIT Profits before interest and taxesPBR Performance-Based Regulation PJM The Pennsylvania, New Jersey and Maryland power pool in the USAPOR Planned Outage RatePSS Power System StabilisersPUC Public Utility CommissionPWR Pressurised Water ReactorsPX California Power ExchangeRoCE Return on capital employedRoE Return-on-equityRoR Rate-of-Return regulationSCA Scheme of Control Agreement(s)SPA Seasonal Power AdjustmentSRMC Short-Run Marginal Costs TSO Transmission System OperatorUK United Kingdom


xxii

1 INTRODUCTION AND OVERVIEW

1.1 Background

Study Team

1.1.1 Environmental Resources Management (ERM) was commissioned by the Government of the Hong Kong Special Administrative Region to undertake a study entitled: Interconnection and Competition in the Hong Kong Electricity Supply Sector. The Study Team comprised experts from the following organisations based in Hong Kong, the United Kingdom, Germany, the United States and Mainland China: ERM Hong Kong ERM Energy (UK) Fichtner GmbH & Co. Public Service Electric & Gas Co. Burns & Roe Co. Tabors, Caramanis & Associates Guangdong Energy Techno-Economic (ETE) Research Centre

Study Overview

1.1.2 The objective of the Study was to establish the feasibility of whether additional interconnection between Hong Kong’s two electricity utilities — joining the CLP Hong Kong Limited (CLP) network in Kowloon with The Hongkong Electric Company (HEC) network on Hong Kong Island — and encouragement of competition in the electricity supply sector would be in the interest of consumers.

Hong Kong Electricity Industry Structure

1.1.3 Hong Kong’s electricity needs are served by two separate electricity utilities, each with vertically integrated generation-transmission-distribution-retailing operations: HEC and CLP,(1) which have been serving Hong Kong since 1890 and 1901 respectively.

Institutional Arrangements

1.1.4 Government and the utilities have entered into Scheme of Control Agreements (SCAs) which regulate profits via a rate-of-return on assets approach. The current agreements are due to expire in 2008.

1(?) CLP has arrangements with several separate companies for the provision of electricity generation. For the purposes of the Scheme of Control Agreement and for this Study, they may be considered to operate as an integrated whole.


1

The Existing Interconnected System

1.1.5 HEC customers’ 1998 peak demand was in the order of 2 300 MW and HEC installed capacity is around 3 300 MW. CLP customers’ 1998 peak demand was around 5 300 MW and CLP installed capacity is about 8 000 MW. The Guangdong Electric Power Holding Company (GPHC) system is over three times the size of the Hong Kong system and the whole interconnected Southern China system is over five times the size of the Hong Kong system. Figure 1.1.a is a diagrammatic representation of the interconnected system.

Figure 1.1.a The Hong Kong and South China Interconnected System Configuration

HEC~ 3 300 MW

Guandong~ 36 000 MW

CLP~ 8 000 MW

Yunnan~ 7 000 MW

Guizhou~ 7 000 MW

Guangxi~ 8 000 MW

8 360+MVA

720MVA

2000MVA

500 MVA

360 MVA

Southern China Hong Kong

Note: The MW values are the installed generation capacities in each area and the MVA values are simply the sum of all installed interconnector circuit ratings, not the transfer capability. For example the transfer capability of the CLP-GPHC (Guangdong) interconnection, is about 5 000 MVA. The transfer capability of the HEC-CLP interconnection is discussed in further detail in Section 2 of the report.

Existing Interconnections

1.1.6 The transmission networks of these two utilities have been interconnected since 1981. The CLP transmission network was interconnected with Southern China in 1979 and this interconnection was greatly strengthened in 1992.

1.1.7 The existing interconnection between Hong Kong Island and Kowloon linking the CLP and HEC systems has a somewhat limited capacity relative to the current size of the two systems.


2

1.1.8 The existing HEC-CLP interconnection is used for emergency support, economy power interchanges, sharing of spinning reserve, and mutual backup of both systems. Its operation is governed by an Interconnection Agreement between the two utilities that was drawn up with some Government involvement, and entered into voluntarily by the utilities. It can be terminated by either utility by providing at least one year’s advance notice in writing.

1.1.9 CLP’s 400 kV transmission network is further connected to the Daya Bay Nuclear Power Station and to the Guangzhou Pumped Storage Plant via the 500 kV network in Guangdong Province. CLP and GPHC are also interconnected at 132/110 kV level as back-up circuits to supply power from CLP to GPHC. The total capacity of the interconnection between CLP and GPHC is almost 5 000 MVA.

1.1.10 The CLP-GPHC interconnection is used to deliver CLP’s share of the nuclear and pumped-storage energy from the Mainland to the CLP system and for bulk energy interchanges from CLP to GPHC. The two systems provide each other with some mutual emergency backup.

1.2 Study Objective

1.2.1 The objective of this Study, as stated in the Terms of Reference, is to: establish whether additional interconnection between Hong Kong’s two electricity supply companies and encouragement of competition in the electricity sector would be in the interests of consumers. The Consultants will be required to provide independent findings and recommendations sufficient to enable the Government to formulate a substantive response to public concern on these issues.

1.2.2 This is a feasibility-level Study. The Terms of Reference ask for answers to two main questions, under the following tasks: interconnection: assessment of the potential costs, benefits and

other implications of increasing interconnection capacity between CLP and HEC;

competition: assessment of the potential for competition in the electricity supply sector, evaluation of alternative market structures and identification of the optimal market structure.

1.3 Study Approach

Study Topics and Related Topics

1.3.1 The two topics of the Study — interconnection and competition — are somewhat inter-related, as the options for future electricity sector competition are determined (among other things) by the


3

wider transmission system, including interconnections between systems.

1.3.2 This Study is not specifically about the following: generation expansion planning; the existing industry structure and ownership arrangements; the advantages and disadvantages of the Scheme of Control

Agreements; possible alternative forms of regulation.

1.3.3 However, it is not possible to consider the possibility of making increased use of interconnection without considering generation expansion planning in substantial detail. Similarly, it is not possible to consider possibilities regarding future competition without considering the existing industry structure and ownership, the existing form of regulation and adjustments, changes or reforms that could be made in future.

Study Constraints

1.3.4 International experience has shown that the process of electricity market restructuring is a long and challenging process, with extensive technical and economic analysis, research and consensus-building required over a significant period of time to accomplish consumer benefits. Hong Kong is no different. The Consultants consider the results of this Study robust, but they should be seen as the beginning of a process and not as the final answer.

1.3.5 This Study is the first of its kind in analysing the electricity market in Hong Kong and identifies issues which will require negotiation and discussion as well as further analysis. Constraints that were encountered in this Study include: the time frame within which to conduct analysis and research

into the complex issues central to the objectives of the Study; some disputed issues on the preciseness of technical and cost

issues concerning increased interconnection between HEC and CLP;

modelling assumptions that had to be made for a 30 year period; the complex policy context of some of the issues; and lack of detailed technical data on the Mainland China electricity

system.

Interpretation of the Results

1.3.6 The quantitative results of this feasibility study are intended to be indicative of the potential economic, financial and tariff benefits — they are not intended to imply formal review and acceptance by


4

Government of such issues as commissioning dates for generating units as projected in the scenarios.

1.3.7 Three time periods are presented in this Study — to 2008, to 2018 and to 2028 — as required by the Terms of Reference. There is naturally greater certainty in the values in the near-term years than in the medium- to long-term years. Costs of generation plant and associated site, fuel supply and transmission costs for instance are more certain in the short than the long term. For example, international prices for combined cycle gas turbine units have fallen substantially in the last few years as the technology matures and the size of the market has increased. This technology was not generally available 30 years ago, and there is considerable uncertainty surrounding the technology that will be available 30 years from now. New technologies, such as small scale fuel cells, are appearing on the horizon and may reach commercialisation well within the time period. Furthermore, there is great uncertainty associated with predicting the effects of major market changes, within and outside the electricity sector in both Hong Kong and China over such a time period. Caution is advised in the interpretation of the medium- to long-term economic, financial and tariff results presented in this Study.

The Current Situation and Context

1.3.8 Both of the questions to be answered by this Study — on increased interconnection and future competition — are inter-related with a number of other broader considerations regarding the current situation of the electricity sector in Hong Kong. The context is therefore crucial to this Study. Relevant aspects of the current situation are outlined below: The SCAs, as they have developed historically, are based on a

model of two separate and independent utilities. The current SCAs are not due to expire until 2008, although there

is a five-year interim review due in 2003. The utilities each plan their own systems then obtain permission

from Government to construct the proposed assets so that they may then count those items in their asset base to obtain a return on them as permitted in the SCAs. The Government does not plan the development of the electricity system, it regulates it.

There exist a given set of investments in generation assets and transmission infrastructure, and one committed but partially completed investment in generation, that are the result of the historical forecasting, planning and approval process.

In power system terms, HEC is small, CLP is somewhat larger and GPHC and the wider Southern China system are very large.


5

Implications of the Current Situation for this Study

1.3.9 Each of these aspects of the actual situation at present has implications for the two questions to be answered by this Study.

1.3.10 The given set of investments in place have an influence on the economic analysis of future investments.

1.3.11 The form of regulation embodied in the SCAs — rate-of-return on fixed assets — has a number of advantages,(1) however one particular disadvantage of this form of regulation is that it discourages the regulated company from seeking least-cost investments in fixed assets, because the permitted return is a function of the company’s fixed assets. This can be managed readily by sound regulation in the case of individual, independent companies, since construction of assets must be approved by the regulator. So the regulatory model — whereby Government has the role of passive approval and the utility has the role of active planner — is well suited to the situation of two separate utilities that developed historically in Hong Kong.

1.3.12 However, these regulatory arrangements tend not to fit so naturally with a situation where a least-cost capital expenditure option would require co-operation between two separate and separately-regulated companies. An example of this is the form of co-ordinated or joint planning that is implicit in consideration of making increased use of interconnection — a topic of this Study. Therefore, the existence of the current SCAs, which are legally binding agreements between Government and commercial entities, may in practice constrain to some extent the options available prior to 2008.

1.3.13 Increasing interconnection between HEC and CLP, would open up the potential in future for meeting some or all of the generation reserve requirements of one area with plant capacity from the other area and/or serving some of the load in one area with generation plants located outside that area. Under the present SCA structure or under possible future competitive electricity market structures, or both, the utility companies may be concerned about this from a commercial perspective.

1.3.14 This may be a greater concern to the smaller HEC system than to the medium-sized CLP system, when only the two Hong Kong utilities are considered, as was the case in this Study, However, CLP is already strongly interconnected with the very large neighbouring GPHC system, and therefore CLP may also have a similar commercial concern with respect to increased use of the CLP-GPHC interconnection.

1(?) Refer to Section 3.3 under Current Alternative Forms of Regulation for the Non-Competitive Parts of the Industry for a discussion of the advantages and disadvantages of various forms of regulation, including rate-of-return regulation.


6

1.3.15 In the short term, increased use of the CLP-GPHC interconnection could, for example, take the form of explicit recognition of the Southern China system for the purposes of determining reserve plant requirements as well as purchase of surplus electricity.(1) In the longer term it would involve exploring the possibilities for introducing competition. Such consideration of competition is discussed in Section 3 of this Report but the opportunity for making increased use of the large, existing CLP-GPHC interconnection for either meeting reserve requirements or for purchase of surplus capacity was not a part of the brief for this Study. It should be borne in mind that the calculations underlying the reserve capacity requirements used to determine the economic benefits of interconnection assume that CLP does not receive any capacity support from Southern China and that sufficient reserve plant must be constructed within Hong Kong(2) to meet Hong Kong’s generation level reliability requirements.

Utility Planning Horizon and Study Time Frame

1.3.16 The utilities currently plan generation and transmission resources to meet expected customer demand for electricity for a period some ten to fifteen years into the future. This planning horizon is typical of most electricity utilities around the world. The options available, particularly in the earlier part of the planning period are developed in detail by the utility planners. Options for the later years are more general or “conceptual.”

1.3.17 As indicated above, the Study brief requires consideration of three time periods — to 2008, to 2018 and to 2028. Table 1.3.a shows the accuracy of the data and information on each of the three main areas considered in this Study — interconnection, competition and the development of the Southern China electricity sector — for each of the three time periods. The quantitative analysis on interconnection in Section 2 and the qualitative analysis of competition in Section 3, should be read bearing in mind the degree of accuracy in the input data and information about the future for each of the three time periods.

1(?) Refer to the discussion of the projected surpluses in Southern China in Section 4.

2(?) The installed capacity considered in the calculation includes 70% of the Daya Bay nuclear plant and 50% of the 1 200 MW Guangdong pumped storage, consistent with CLP’s contractual entitlements.


7

Table 1.3.a Accuracy of Data and Information in Each Period of the Study Time-frame

Time Period Information on generation and interconnection RESOURCE OPTIONS

Development of COMPETITION

Physical and institutional development of Southern CHINA electricity sector

to 2008 detailed, including sites and quite detailed engineering cost estimates

options for initial steps possible but limited by practical considerations

reasonable predictions can be made

to 2018 generic and more approximate, particularly towards the end of the period

options for reform quite open if planned in advance

predictions are quite uncertain

to 2028 must be considered very approximate

options for reform very open if planned well in advance

predictions are extremely uncertain

Interconnection

Technical and Economic Analysis of Increased HEC-CLP Interconnection

1.3.18 An assessment of the overall economic benefits is necessary to calculate the potential consumer benefits from increased interconnection between the Hong Kong utilities. To achieve this, it is logically necessary to calculate the economics of alternative resource options from an overall least-cost system planning approach.

1.3.19 Sources of economic benefits include: reducing the overall reserve plant required in the interconnected

system, such that the additional capital cost of increased interconnection is more than offset by the savings in generation capital cost;

increasing the degree of flexibility available for selecting future plant sites; and

benefits whereby greater flexibility of generation plant dispatch can allow overall lower cost system operation.

1.3.20 The first step in such an exercise is to identify the alternative resource options available to meet expected (or “forecast”) customer demand for electricity with the desired level of reliability.


8

(1) This includes integrated consideration of generation and transmission (in this case specifically increased use of HEC-CLP interconnection) resource options. This allows comparison between several scenarios. Each scenario in this Study is based on common information on projected customer peak demand and energy consumption in each utility area for the Study period. Under this approach, the potential economic benefits of each particular investment scenario is determined by comparison with the base case investment scenario. The total benefits of the overall least-cost option are then simply the difference between the present value of its cost stream and that of the base case.

Resource Options Considered in the Scenarios

1.3.21 The specific near-term generation resource options developed by the utilities’ planners were used in this Study but the brief for this Study required analysis up to 2028 — which extends well beyond the utilities’ planning horizon — so additional options had to be sketched out for the later years required by the brief.

1.3.22 The generation resource options considered for the near term included: the last two single shaft 312.5 MW combined cycle gas turbine

(CCGT) units at the Black Point power station (for which the civil works and transmission connections are already in place) fuelled by the existing piped natural gas supply;(2) and

six single shaft combined cycle gas turbine (CCGT) units of approximately 300 MW at the proposed reclaimed land extension to the Lamma power station fuelled by natural gas supplied by a proposed pipeline from a proposed LNG terminal in southern China.

1.3.23 For the period after this, the conceptual plans of one utility include some small gas turbine units to meet peak load requirements and some 660 MW conventional coal-fired units. This Study however, was based on the following generation configurations to be the scenario for this period: several small gas turbine units on available space at the Black

Point site; and

1(?) In comparing resource options and the costs of various scenarios, it is important that the reliability level specified be the same between the various cases, to enable the economic feasibility of interconnection to be isolated from the effects of changed reliability levels. This Study endeavoured to specify reliability levels the same as those currently applied by the utility planners in Hong Kong to model the various scenarios. However, there are some differences between the two utilities with respect to reliability criteria and the analytical methodology currently applied by system planners. There was a logical necessity to harmonise reliability criteria to the extent possible for the purposes of the economic feasibility analysis in this Study. This Study notes that a review of system planning and reliability criteria may be appropriate in Hong Kong, although the results of such a review would be very unlikely to affect the overall conclusions of this Study. 2(?) CLP has contractual commitments, based on agreements with Government, to purchase these two units.


9

future generic combined cycle units,(1) with large 680 MW triple-shaft CCGTs instead of coal-fired steam plant, in line with the current Government thinking that there will be a move away from coal-fired electricity generation in favour of lower emission gas-fired plant for environmental reasons.

Financial and Tariff Analysis of Each Scenario

1.3.24 The manner in which the overall economic benefits flow through to consumers depends on the way that revenue requirements (including profits) and tariffs are determined. Hong Kong electricity revenues and tariffs are determined by the Scheme of Control. The investment schedules of each interconnection/generation scenario for each electricity company from the economic analysis can therefore be used together with the formulae in the Scheme of Control accounting system to calculate the aggregate effect on customer bills and the average effect on customer tariffs of each scenario relative to the base case.

Considerations of Ownership Structure

1.3.25 The technical and economic analysis of least cost interconnection/generation scenarios is done from an overall integrated system or total resource point of view. Such consideration is quite distinct from questions of the division of asset ownership between areas within the interconnected electricity supply system. Nevertheless, such considerations are important, and were considered by this Study. Considerations related to ownership include: the effect on the management and control of each utility system; each individual company’s ability to serve reliably the load of

customers within its supply area and the extent to which it needs to rely on inter-area generation capacity support and energy supply from a separate company in a neighbouring area; and

the effect of least cost investment options on the commercial and financial position of each individual company in the interconnected system.

1.3.26 These considerations are discussed in this report in the context of the extent to which the existing Scheme of Control regulatory framework and the Interconnection Agreement would require adjustments or modifications to those agreements to permit solutions to be found to these issues.

1(?) The first 680 MW unit could be accommodated within available site space, but subsequent units would require formation of additional space at the Black Point site or creation of a new site. Sufficient transmission capacity exists at Black Point to accommodate such additional units up to and including the first 680 MW combined cycle gas turbine. Units beyond that would require construction of additional transmission capacity.


10

Competition

1.3.27 Examination of the potential for the future introduction of electricity sector competition allows the wider implications of making increased use of the HEC-CLP interconnection to be considered, including future development of the industry, and the future relationship between the electricity industry in Hong Kong and the rapidly developing Southern China system, with which Hong Kong is already strongly interconnected.

1.3.28 After the technical and economic analysis of interconnection from the perspective of the overall Hong Kong system, the Study report focuses on the potential for introduction of competition. It: examines the local and wider context; describes alternative electricity market structures; evaluates in detail the most appropriate options for Hong Kong,

indicating a set of options and possible time-frames for transition from the existing structure.

1.4 Structure of this Document

1.4.1 This document is organised into Sections as follows.

1.4.2 Section 2: Feasibility of Increased Interconnection of Electricity Systems in Hong Kong – including: general considerations associated with electricity interconnection; methodology and approach adopted to determine the technical

and economic feasibility of increased interconnection between HEC and CLP;

an analysis of the technical feasibility of increased HEC-CLP interconnection; and

an analysis of the economic feasibility of increased HEC-CLP interconnection.

1.4.3 Section 3: Feasibility of the Introduction of Electricity Sector Competition in Hong Kong – including: local and wider context for electricity sector competition; evaluation of four market structures applied to Hong Kong; and detailed consideration of the most appropriate options for

Hong Kong.

1.4.4 Section 4: Developments in the Southern China Power Markets – including: Guangdong and Southern China electricity sector; process of restructuring; and implications for Hong Kong.


11

1.5 Key Terms

1.5.1 To ensure clarity and avoid confusion when describing the various options and case studies analysed, a description of terms used in the report is detailed in the glossary at the beginning of the document. Specific terms used in the tasks and modelling are included in Table 1.5.a.

Table 1.5.a Key Terms Used in this Study

Topic Key Term Description

Planning and Operation of Interconnection

Joint Planning Planning generation and transmission taking into account interconnection options in order to achieve least-cost expansion of the system as a whole.

Firm Capacity Joint planning would require that the interconnection be able to provide “firm” inter-area generation capacity support as and when it is needed. The use of the term “firm capacity” in this report refers to the technical capability of the interconnection itself. It does not mean that generation capacity in one area is set aside or reserved for the sole or preferential use of the other area.

Joint Dispatch Operating or “dispatching” generation plants in separate areas on a joint basis to realise any available synergy and operational economies to achieve the least cost generation production cost.

Firm Power Joint dispatch of plant would require that the interconnection be able to support the regular, reliable transfer of energy. The use of the term “firm power” in this report refers to the technical capability of the interconnection itself. It does not necessarily refer to contractual arrangements between utilities for the supply of energy, although neither does it preclude such arrangements being put in place.

Technical and Economic Analysis of Interconnection

4 Interconnection Cases

The four interconnection cases were developed as generation/interconnection scenarios to examine in detail the potential for increased use of interconnection to benefit consumers.

Financial and Tariff Analysis of Interconnection Cases

Financial Model

Taking the recommendations from Task 1, the Financial and Tariff Analysis provides an indication of the average effect on electricity consumer tariffs of each of the interconnection cases under the financial arrangements of the existing Scheme of Control.

Competition 4 Market Models each with several variants

A wider analysis of options to introduce competition into the Hong Kong electricity market.


12

2 FEASIBILITY OF INCREASED INTERCONNECTION OF ELECTRICITY SYSTEMS IN HONG KONG

2.1 Introduction

2.1.1 World-wide there is a trend towards increasing interconnection between utility systems. This trend is in response to customer demand for delivery of competitively priced electricity to their area as well as helping to facilitate the introduction of competition in the electricity industry. The decision to introduce competition is often based on a general conviction among policy makers that competitive markets tend to deliver the best solutions for customers, offer a means of ensuring that the optimum amount of generating capacity is provided with minimal or no Government intervention and provide a framework that naturally encourages economic exchange of energy and sharing of capacity requirements between systems. In implementing decisions on interconnection, utilities look for associated benefits in reducing generating capacity, providing emergency support, improving security of supply, reducing fuel costs, gaining economies of scale and enhancing the possibilities for increased competition in the longer term, where their market position may lend an advantage. This section focuses on the technical and economic aspects of the actual interconnection itself between the HEC and CLP systems.

This Section

2.1.2 The interconnection of electricity systems is important in helping to ensure security of supply, the provision of economic benefits to consumers, the facilitation of possible competition in future and many other benefits. Given these considerations, this Section discusses: general considerations associated with interconnection; the methodology and approach adopted to assess the technical

feasibility of increased HEC-CLP interconnection; the technical feasibility of increased interconnection between

HEC and CLP; and the economic feasibility of increased interconnection between

HEC and CLP, including an assessment of the overall economic benefit to Hong Kong and the consequent effect on average tariffs and total consumer bills (or conversely on utility revenues) under the current Scheme of Control Agreements between Government and the utilities, of the economic benefits from increased interconnection.


13

2.2 General Considerations Associated with Electricity Interconnection

What is Interconnection?

2.2.1 Interconnection is simply the linking of two electricity systems to tie together the major electric system facilities, generation resources and customer demand centres of the previously separate systems.

2.2.2 The systems of the two electricity companies in Hong Kong — HEC and CLP — are already interconnected by a submarine cable link of relatively limited capacity. Because the HEC system is on an island, any future upgrades to the interconnection capacity would also need to be by submarine cables.

2.2.3 Interconnecting transmission systems is one of the principal ways of achieving a reliable electricity supply. Like all transmission systems, interconnected systems must be planned, designed and constructed to operate reliably within thermal, voltage and stability limits, while: delivering electric power to areas of customer demand; providing flexibility for changing system conditions; allowing for reduced reserve costs; allowing the amount of installed generating capacity required to

meet the specified reliability level to be lower than would otherwise be the case; and

providing for economic exchange of electric power among systems.

Potential Benefits from Interconnection

2.2.4 Realising potential economic benefits from interconnection requires that a system be planned or operated or both planned and operated to take maximum advantage of that interconnection. The former is referred to in this report as Joint Planning and the latter as Joint Dispatch. The maximum technical and economic benefits of interconnection, potentially available from optimising the generation and transmission configuration of multiple areas that make up an interconnected system and from operating or “dispatching” the generation plant in the most economic or least cost way can be determined independently of the ownership structure of the individual utilities within the interconnected system.(1) The implications of the particular commercial realities of a given system — including the ownership structure and regulatory framework — must be considered in addition to the technical and economic analysis of the potential benefits of interconnection. This feasibility

1(?) This feasibility Study focuses primarily on the interconnected HEC-CLP system — there may be additional potential economic benefits available in future from taking full advantage of the interconnection with the Southern China


14

Study does so, taking into account technical, economic, consumer and utility commercial considerations in turn.

Modes of Operation of Interconnection

2.2.5 The approach to planning and/or operation of interconnected systems requires consideration of the alternative modes of operation of the interconnection between those systems. The benefits available from interconnection depend on these modes of operation, and the extent to which they are reflected in system planning.

2.2.6 The main modes of interconnection operation are referred to in this report as: Firm Capacity, necessary for Joint Planning; Firm Power; necessary for Joint Dispatch; and the combination of Firm Capacity and Firm Power, necessary for

the combination of Joint Planning and Joint Dispatch. The combination of Firm Capacity and Firm Power is commonly employed internationally.

2.2.7 These modes represent two basic characteristics of electricity systems whereby demand fluctuates continuously, but supply and demand must at all times be equal. The two characteristics are: capacity: the need to have sufficient generation capacity to meet

customer demand at its maximum (peak demand) and a transmission system of sufficient capacity to meet that demand as it is distributed throughout the network; and

energy (referred to here as “power”): the requirement to supply energy.

2.2.8 Interconnection between two or more separate systems allows greater flexibility in the planning of additional generating units, reduced reserve margins, and the other benefits discussed later in this section, by utilising the interconnector under ‘Firm Capacity’. The situation may arise where no power at all is passed across the interconnector and the benefits are from the rationalisation of investment in new generating units. In such cases, benefits usually arise from reduced reserve margins, reduced spinning reserve, emergency support and security of supply (these issues are further described below and represented in Figure 2.2.a). However, Firm Capacity is not defined as one type of activity, rather it is a series of activities, some of which are already practised here in Hong Kong. See Table 2.2.a for a description of the spectrum of activities that would be considered under Firm Capacity (all issues summarised in the table are discussed in full later).


15

Firm Capacity

2.2.9 Under the Firm Capacity mode of interconnection operation, the interconnector is used to provide mutual capacity support between the interconnected utilities. Flows of energy are very small, because they occur only for relatively short periods on those infrequent occasions when backup support is needed by one or other utility. Table 2.2.a shows the various potential uses of an interconnector under the Firm Capacity mode of operation.

Table 2.2.a Spectrum of Benefits available from Considering Firm Capacity of Interconnection

Potential Use Description

Increased Level I (Generation) and Level II (Transmission) Reliability

Increasing the level of interconnection within and between electricity systems increases their reliability, as it increases the number of possible combinations of generation sources and transmission network paths that can be used to supply load, which reduces the probability that load cannot be served in the event of a fault at any particular point in the system. Hong Kong already experiences some of these benefits to the extent allowed by the capacity of the existing interconnection.

Sharing of Spinning Reserve and Load-Frequency Dependence

The next step is to share spinning reserve, the sum of the spare capacity in spinning generation units available for instantaneous support in the event of a generator being “tripped” (unexpectedly forced out of service). Interconnection allows spinning reserve to be shared among the interconnected parts of the wider system and also provides advantages from the frequency-dependence of load. Both of these types of benefit are currently captured in Hong Kong, where spinning reserve is currently shared between HEC, CLP and GPHC on the wider interconnected system and where the frequency-dependence of load across the wider interconnected system means that the maximum spinning reserve that needs to be carried is just 75% of the largest unit on the wider interconnected system. This consideration and the calculation that shows how the 75% is arrived at are explained later in this Section.

Co-ordinated or joint generating capacity expansion planning

By interconnecting with a neighbouring utility, installed reserve generating capacity can be shared such that less generation is required in total to meet the desired generation reliability level than would be the case if the two systems were not interconnected, or planned capacity expansion as if they were not interconnected. While the Hong Kong utilities take into account the reliability benefits of reserve plant in the adjoining interconnected systems to some extent, Joint Capacity expansion planning to the full extent possible, that would minimise generation and transmission costs for the overall system could be further exploited.(1)

1(?) The extent to which Joint Planning is currently practised is described below in sub-section 2.3 Methodology and Approach under Planning for Electricity System Reliability, Overall Adequacy of Generation. Extensions to current practice include co-operative examination of interconnection options that would provide the least-cost way of meeting projected loads with the appropriate level of reliability and including in Hong Kong system planning the effect of generation support from Southern China via the CLP-GPHC interconnection. The former is examined in detail in this feasibility Study on increased use of HEC-CLP interconnection. The latter is noted as a potential future improvement, which would require an assessment of the


16

Firm Power

2.2.10 Under the Firm Power mode of interconnection operation, the interconnector is used for the purpose of energy exchange between the systems. Under Firm Power, transfers of energy occur on a regular basis. There are a variety of possible institutional arrangements under which such Firm Power flow may occur, including simple economy transfers, bilateral contracts or power pooling. Table 2.2.b shows the various potential uses of an interconnector under the Firm Power mode of operation.

Table 2.2.bSpectrum of Benefits Available from Considering Firm Power Transfer Across Interconnection

Potential Use DescriptionEconomy Exchange

As has been practised in Hong Kong, this is a very limited form of voluntary co-ordinated dispatch of generating units.

Power Purchasing Contracts

This is a situation whereby one utility may contract to purchase power from another utility over a set period of time, ranging from hours to days or even years. The current Interconnection Agreement between HEC and CLP has limited provisions for power purchases.

Joint Dispatch The situation where all generating units in two or more systems are dispatched — generally by one organisation that may or may not be jointly owned by the utilities involved — according to agreed rules. This is not currently practised in Hong Kong.

Potential Benefits of Interconnection

2.2.11 The potential uses of interconnection under Firm Capacity and Firm Power interconnection arrangements provide benefits. These include: investment savings from simple reduced reserve capacity

requirements, economies of scale in unit size selection and somewhat more complex considerations whereby greater flexibility may be achieved in the selection and planning of additional generating units; and

operational savings whereby it is possible to operate the generation units on the interconnected system in a way that is less expensive than operating the two systems separately.

These general potential benefits are indicated in Figure 2.2.a. Most utilities world-wide achieve the maximum benefits from the implementation of both Firm Capacity and Firm Power.

load and demand parameters of generation units in the Southern China system and effect of transmission constraints in that system on the ability of those generation units to support loss of load in Hong Kong.


17

Figure 2.2.a Potential Overall Benefits of Interconnected Systems

OperationalSavings

InvestmentSavings

System 1 System 2

Peak load

Load Pattern

Installed Capacity

Unit Forced Outages

Reserve Requirements

Fuel Diversity ofGeneration Units

Peak load

Load Pattern

Installed Capacity

Unit Forced Outages

Reserve Requirements

Fuel Diversity ofGeneration Units

Increased EconomicTransfers

Decreased ReserveRequirements

2.3 Methodology and Approach

General Approach Adopted to Assess Potential Benefits

2.3.1 The Consultants were asked to undertake a feasibility-level study of the benefits and costs of increased interconnection in Hong Kong. This is not a generation expansion planning exercise as such. However, the only way to assess the economic benefits of increased interconnection is to develop scenarios that combine generation expansion planning under alternative interconnection cases.

2.3.2 Therefore, the general approach adopted to assess the potential benefits of increased interconnection was broadly as follows: identify alternative interconnection upgrade options; develop a base case generation expansion scenario for the case

with the existing level of interconnection, taking into account a specified set of planning criteria;

assess by inspection of plant installation dates and costs, and systematic steps, the most appropriate installation dates of each of the alternative interconnection upgrade options identified;

define the best candidates among these as cases; develop generation expansion plans for each interconnection case

to form an interconnection/generation scenario, taking into account the specified set of planning criteria;

use the capital expenditures over time on each of the transmission and generation installations to calculate discounted net present values for each scenario;

assess the differences in net present value between each alternative scenario and the base case scenario for the Firm Capacity economic analysis.


18

Both this general approach and the detailed analytical method adopted ensure that reliability for each individual area (HEC and CLP) would be equivalent for the generation expansion plans under each of the various generation/interconnection scenarios, by applying a common and consistent set of reliability criteria. Reliability considerations in general and the particular reliability criteria and their application for this Study are discussed below.

Planning for Electricity System Reliability

2.3.3 Electricity is supplied to customers by a system that extends from the generation plant through a high voltage transmission network and finally a low voltage distribution network to the customers’ premises. An interruption anywhere in this chain of supply to a particular customer or group of customers will result in the loss of supply or “loss of load” for those customers.

2.3.4 Clearly, problems further up the network have the potential to affect large numbers of customers unless sufficient redundancy is built into the system. For this reason, electricity system reliability planning needs to consider overall Level I (generation) and Level II (transmission) reliability to determine the optimum amount and mix of reserve generation, its location within the transmission network, and the optimum configuration and degree of redundancy in the transmission network necessary to ensure the desired level of reliability. Level III (distribution) reliability represents a more detailed level of planning and is less critical from an overall system point of view, as distribution level faults affect only local areas.

Generation (Level I) Reliability

2.3.5 Generation level reliability is assessed by ensuring that for the expected level of hourly load throughout each day, week, month and year of the planning period there will be an adequate amount of generation plant installed overall to allow for: units to be taken out for maintenance; sudden and unpredictable unavailability of generation units; and the ability to cope with unpredicted peak loads. Maintenance is performed at the low demand times of the year, and is known as “planned outage,” expressed in terms of a planned outage rate POR, in percentage of time or weeks per year. Sudden loss of units — due to units being “tripped” by the automatic protection equipment that is built into the generation and transmission system to ensure that it operates safely — is known as “forced outage” and is expressed in terms of a forced outage rate or FOR, in percentage of time.

2.3.6 Two tests are applied to assess the required levels of installed generation:


19

a probabilistic test, that assesses the generation units not on maintenance at each time of the year against the expected load at each time of year, then finds the combinations of plant outages that would result in available generation being less than the load in each time period and uses the FOR (probability) values for each plant to calculate the overall “loss of load probability” or LOLP, which is then compared to a specified LOLP criterion; and

an operational test, that checks whether for each year of the planning period there would be sufficient installed capacity to allow the peak load to be served under a specified contingency.

Overall Adequacy of Generation

2.3.7 The specified LOLP or “primary” criterion currently used by HEC for system planning is 20 hours per year. The specified LOLP or “primary” criterion currently used by CLP for system planning is 0.5 days/year.

2.3.8 Before construction of the interconnector, HEC applied an LOLP criterion of 2.5 hours/year. After construction of the existing interconnection, HEC continued to model their system as an isolated system, but relaxed their LOLP criterion to 20 hours/year, and to take its existence into account continued to model their system as an isolated system. HEC calculated that the result of this approach would be approximately the same as continuing to use a 2.5 hours/year LOLP criterion and conducting a two-area LOLP analysis of the interconnected HEC-CLP system.(1) However, the approximation made in the adjustment to the criterion allows the continued application of a single-area LOLP approach, rather than requiring application of the more complex two-area LOLP approach.

2.3.9 CLP apply their 0.5 days/year LOLP criterion to their system, taking into account the existing interconnection with HEC; including their share of Daya Bay nuclear plant capacity and Guangdong pumped storage; but excluding the interconnection of the CLP network with the wider Southern China grid. The existence of the interconnection with the Southern China grid means that the level of generation reliability experienced by the CLP system is in fact higher (lower LOLP) than the calculated level. In other words, using 0.5 days/year and not taking account of the interconnection with the Southern China grid would be equivalent to using a lower LOLP criterion and explicitly taking account of support via the interconnection with the much larger Southern China system.

Allowance for Generation Contingencies

2.3.10 As well as having an adequate total level of installed generation capacity to keep the probability of loss of load within the specified

1(?) Note that single-area to two-area equivalent LOLP values would change as the system changes, so the value used is quite approximate. This Study adopted a full two-area LOLP approach, which obviates the need for making a single-area equivalence approximation.


20

limit, the generation system must be able to continue to serve load in the event of specified contingencies. It is only necessary to specify the worst case generation contingency, which is represented by the loss of the largest unit or units. Once this contingency can be met, all other smaller generation contingencies will automatically be met.

2.3.11 The specified generation level operational contingency or “secondary criteria” currently applied differ between the two utilities. They are: HEC Reserve Capacity > Largest Unit + Hot Standby(1) CLP Installed Capacity > Max Demand + 2 x Largest Unit +

Spinning Reserve(2)

Transmission (Level II) Reliability

2.3.12 Transmission level reliability analysis assesses the reliability of the energy transport system. It requires consideration of the distribution of loads throughout the network (substation loads), rather than simply the total system load as used in generation level reliability assessment and planning. It applies methods which examine the ability of the system to remain stable and serve load given the occurrence of faults or “contingencies” in the network. If, for example, a single contingency or N-1 rule is applied, then the failure of each single transmission network element is examined in turn to determine whether there is sufficient redundant capacity in the network to continue to serve load under all planning contingencies, or whether additions or modifications to the network configuration are required. More precise methods also require consideration of all relevant system stability effects such as transient stability as well as system oscillations.

Application of Reliability Analysis for this Feasibility Study

Generation

2.3.13 A consistent and systematic approach to generation planning criteria was required for this study across all scenarios examined to allow the economic benefits of deferral of capital expenditure of plant due to the interconnector to be clearly identified.

2.3.14 The following approach to generation planning criteria was applied for this feasibility study across all scenarios examined to allow the

1(?) Hot Standby is the commitment to provide further emergency support after all spinning reserve is utilised. The current value for this is 180 MW in the HEC system. 2(?) Not all generating plants in a power system are operating at their full output capacity all of the time. Spinning reserve is the sum across the system of reserve capacity (rated output capacity minus current output) in units that are “spinning” or currently generating electricity. Spinning reserve is shared between HEC, CLP and GPHC. CLP’s commitment to the system is currently 240 MW.


21

economic benefits of deferral of capital expenditure of plant due to the interconnector to be clearly identified: a primary LOLP criterion of 0.5 days/year, for both HEC and CLP,

calculated on a two-area basis, taking into account the total transfer capability of the interconnector; and

secondary operational criteria of: HEC Installed Capacity > Max Demand + 2 Largest Units(1) CLP Installed Capacity > Max Demand + 2 x Largest Unit(2) +

Spinning Reservetaking into account capacity according to whichever was the smaller of reserve capacity(3) in the supporting area and the available transfer capability of the interconnector.

2.3.15 The application of the primary(4) and secondary criteria(5) as defined above would ensure that the generation level reliability for each of the two utilities would not fall below the reliability as specified by the present criteria.(6) The application of the two-area LOLP method

1(?) The largest units in the HEC system are currently 350MW. Two of these will be re-powered to 375MW in 2001 and 2002. The largest unit value was adjusted accordingly in those years. 2(?) The two largest units in the CLP system are currently taken as 690MW, which is the CLP share of the two Daya Bay nuclear units’ capacity. 3(?) Reserve capacity in the supporting area equals installed capacity in the supporting area minus peak demand in the supporting area. 4(?) HEC have indicated that 0.1 days per year would be a more appropriate LOLP criterion for Hong Kong. A consistent criterion was needed for the purposes of developing scenarios for this feasibility Study. The selection of 0.5 days per year LOLP criterion is conservative from the point of view of estimating potential economic benefits from increased use of interconnection (which was the purpose of this feasibility Study). In other words, if increased interconnection is economically feasible using 0.5 days/year, then it will also be feasible, but with greater potential benefits using LOLP 0.1 days/year, because adopting an LOLP criterion of 0.1 days/year would produce more generation plants in all scenarios and would thus result in higher capital expenditure deferral benefits between scenarios than are shown in this Study. However, from the point of view of reliability, adopting an LOLP criterion of 0.1 days/year is more conservative than the 0.5 days/year adopted in this Study. Refer to the discussion below on Observations on Current Planning Criteria. 5(?) Note that the secondary criteria are here applied jointly to the interconnected system, rather than independently. At least one utility is of the view that their system should be planned by the independent application of the two largest units secondary criterion to their system, based on the premise that, until 2008 under the SCA, their company is ultimately responsible for the reliability of its system and therefore cannot be expected to rely on reserve generation capacity from another area under the responsibility of another company. The Consultants acknowledge this concern as a regulatory and institutional issue and have considered it in the context of the lack of “fit” between the pursuit of overall economic benefits potentially available from increased interconnection and the present Scheme of Control and Interconnection Agreements. Applying the two largest units criteria to each area independently would result in a higher level of installed capacity and hence higher generation level reliability than would its application to the system as a whole. Government may wish to consider whether such a cautious approach is appropriate in the interim period before a full review has (a) established whether or not a more stringent LOLP criterion than 0.5 days/year is appropriate for Hong Kong, (b) decided what is the most appropriate methodology for considering inter-area support between CLP and GPHC as well as between HEC and CLP now and in the future in system planning, and (c) compiled a full-scope review of Level I and Level II reliability for the overall interconnected generation and transmission system. 6(?) Note that percent reserve margin is an output from an analysis using the primary and secondary criteria described here, rather than an input to the analysis. The Consultants consider that, while simple percent reserve margin indices provide a quick and convenient way of comparing reserve levels in different systems or at different times in the same system, the application of such a parameter does not make for a rigorous engineering approach to planning


22

in the calculation of overall generation adequacy is important in this respect.

2.3.16 The generation installation scenarios under the various interconnection scenarios were developed using a two-area LOLP analysis(1) as follows: sufficient generation plant must be installed to allow both the

primary probabilistic criterion and the secondary operational criterion to be satisfied in each individual area in each year of the analysis period;

when local capacity in one area is unable to serve the local load in its area (as may occur in the event of outage/s), support may be obtained from the other area, with this support constrained by the N-1 (single transmission contingency) transfer capability of interconnection between the areas; but

such support from the other (ie supporting) area can only come from plant not at that time serving load in the supporting area (ie there may be sharing of capacity between the supporting and supported areas, but there would be no sharing of loss of load — the loss of load would be confined to the area that experienced the outage); and

while including the probability of support from the supporting area in the calculation of LOLP for the supported area, the following considerations with respect to the supporting area are taken into account: the load at the relevant time, the installed capacity, the maintenance schedule (planned outages) and the probability of a forced (unplanned) outage in the supporting area simultaneous with the outage/s in the supported area.(2)

2.3.17 The timing of generation plant installation for each interconnection scenario was determined to meet the specified LOLP criterion of 0.5 days/year (LOLE 2 years/day) at the lowest overall present value cost, while also meeting the secondary criterion described above. The capacity provided for under the secondary criterion could come from either area, subject to the interconnection constraint, where that constraint was calculated conservatively by deducting the spinning reserve obligation in each direction from the N-1 (single transmission contingency) transfer capability to obtain the available

the appropriate level of generation adequacy because: percent reserve margin does not and cannot take into account the probability of loss of load, the effect of the maintenance schedule on loss of load probability, the effect of forced outage rates of the various plants in the system, the relationship between the size of the peak load and the size of generation units within the system (it only includes the total capacity in MW in the calculation) and the distribution of loads throughout the year (it only includes the one hour of peak load in MW in the calculation).1(?) The GEBGE multi-area LOLP software (which takes its name from the original developers, General Electric/Baltimore Gas and Electric) was used for this task. This software has a long track-record of use in multi-area LOLP analyses for interconnected utilities in the north-east of the United States, such as those in the Pennsylvania, New Jersey, Maryland or “PJM” pool. The model runs were conducted by Public Service Electric & Gas (PSE&G), an organisation experienced in the use of GEBGE and a member of the PJM pool. 2(?) These are considered in addition to the same factors in the supported area: load, installed capacity, maintenance schedule (planned outages) and the probability of a forced (unplanned) outage.


23

transfer capability.(1) Under this approach, additional generation capacity would not necessarily need to be installed in the same area as the load which is serves, as long as both the LOLP criterion and the operational criterion were able to be met in each area.(2)

Transmission

2.3.18 As explained above, under Planning for Electricity System Reliability, overall reliability requires consideration of the entire supply chain, including transmission. Increasing generation capacity to increase Level I generation reliability does not increase Level II transmission reliability. A full quantitative transmission-level reliability analysis was not undertaken in this feasibility Study. However, it is observed that increased interconnection between HEC and CLP, particularly via an additional, separate higher voltage circuit would likely increase Level II (transmission) reliability, particularly for HEC, just as the transmission circuits planned to deliver output from future generation additions at Lamma Island would increase Level II reliability. This would be due to the new transmission circuits being electrically separate from the existing circuits. Generation and transmission corridors are reasonably highly concentrated in Hong Kong. A review of Level I and Level II reliability for the overall interconnected system would clarify the implications of the increased concentration of generation at single sites in Hong Kong.

Observations on Current Planning Parameters

2.3.19 This study used existing planning parameters where possible, and only adjusted them where necessary to ensure consistency in modelling a new (joint) approach to future planning. However, several points may be observed about the current planning parameters, which the Consultants recommend should be reviewed, particularly in moving towards a new, integrated approach to electricity system planning in Hong Kong.

Overall Level I (Generation) and Level II (Transmission) Reliability

2.3.20 This feasibility Study involved an analysis of Level I (generation) reliability. However, as explained above, reliability of supply involves the whole supply chain, not just the top of the chain. A comprehensive, system-wide, integrated review of Level I and

1(?) The method described here would represent a change from the current approach to generation planning whereby each utility plans independently, taking into account inter-area support only in a limited way. It is noted that the scope of this study was confined to HEC-CLP interconnection. However, since CLP is strongly interconnected with Southern China and can obtain generation support from Southern China, the actual LOLP values, particularly for CLP, would be somewhat lower (more favourable) than those calculated in this Study by the method outlined here. Joint Planning of generation between Hong Kong and Southern China represents a further possible future development of the method proposed here. 2(?) The Consultants note that — while there are no technical barriers in principle to the load of one area being served by plant from another area in the absence of a generation contingency — planning for such an arrangement would require the Interconnection Agreement between HEC and CLP to be re-negotiated.


24

Level II reliability analysis, taking into account possible future interconnection capacity increases would also cover: transmission reliability, including an analysis of transient effects

such as inter-area oscillations; and substation reliability. Such a review would be advisable to ensure that the overall reliability of supply in Hong Kong is not compromised.

LOLP Criterion

2.3.21 During the Study there was some discussion between the Consultants, Government and the utilities about the most appropriate LOLP criterion to use for Hong Kong. While 0.5 days/year was adopted in this feasibility Study for the purposes of estimating the economic benefits potentially available from increased use of HEC-CLP interconnection, it is clear that if Government decides to adopt a policy of increased interconnection for the Hong Kong electricity sector in the short-to-medium term that LOLP criteria and the methodology by which it is applied should be reviewed.(1) If Joint Planning is eventually adopted in Hong Kong, a consistent approach would be needed to the application of the LOLP criterion. It would also be important to resolve the question of how generation support (in both directions) via CLP’s interconnection with Southern China should be taken into account in the planning process, which would likely involve extensive co-operation on technical and non-technical issues between the Hong Kong Government and electricity companies and relevant organisations in Southern China. Resolving such issues, which may take some time, would allow the current approach to be replaced with an approach that explicitly accounts for all contributions to generation reliability in planning in a way that accurately reflects the configuration and operation of the system. So, if Joint Planning is eventually adopted, and as the Southern China electricity system grows and develops and reliability benefits of the interconnection between CLP and GPHC become more significant, it will become less and less appropriate for CLP to plan its generation system without considering support from Southern China,(2) just as it would at that 1(?) In the course of the Study, examples were provided to the Consultants of other major cities that use an LOLP criterion of 0.1 days/year. It was not properly the role of this Study to review LOLP and other planning criteria. However, several general points may be observed with respect to a possible review of planning criteria. Firstly, reliability is always a trade-off against cost. The optimum trade-off point is found by taking into account what is known as the “cost of unserved load” — the cost to utility customers of supply interruption. A study of these costs for each group of customers in Hong Kong could be used to develop the most appropriate LOLP criterion. Secondly, as explained above, reliability should be assessed comprehensively. So, increasing generation reliability via a more stringent LOLP criterion will only increase overall reliability if transmission (and ultimately distribution) are similarly reliable. 2(?) CLP have stated that they consider that it is, at present, prudent planning practice not to include generation capacity support from the Southern China system in their LOLP calculation due to uncertainty regarding load and demand parameters and unreliability of generation units in the Southern China system and the effect of transmission constraints in that system on the ability of those generation units to support loss of load in Hong Kong. It was not the role of this particular feasibility Study to analyse these issues directly. However, just as a review of the most appropriate LOLP criterion for Hong Kong may be appropriate at this time, it may also be


25

time no longer be appropriate for HEC to continue to plan using the simple adjusted LOLP criterion and a single-area LOLP approach that does not explicitly reflect the generation support available in practice from CLP.

Forced Outage Rates

2.3.22 LOLP analysis results are very sensitive to the Forced Outage Rate input values used. It is advisable for a least cost approach to generation system planning to use FOR values that represent the forced outage rates experienced for plants. Some care needs to be exercised, as the forced outage rates actually experienced tend to be high in the early years after plant commissioning and then “mature” to lower, stable levels once the initial problems with the operation of each particular plant are resolved. Continuing to use high FOR values for system planning LOLP analysis based on the results experienced in the early years of plant life will tend to return results that indicate that more generation than is really required. This feasibility Study used the FOR rates currently adopted by the utilities with Government approval. However, a review of the approach to determining forced outage rates and the actual rates for each generation unit in Hong Kong as used for LOLP calculations may be appropriate.

Secondary Criterion

2.3.23 The existing secondary criteria differ between the two utilities. HEC have suggested that their secondary criterion should include allowance for the loss of two largest units rather than one plus hot standby, and that all of the HEC reserve generation capacity should be located within the HEC system. As described above under the explanation of the application of reliability analysis in this study, this study has adopted an HEC secondary criterion to ensure that there would always be sufficient reserve generation capacity support to cover the loss of HEC’s two largest units. The adopted criterion differs from HEC’s proposed criterion with respect to the location of the reserve generation capacity by considering both reserve generation capacity within HEC and reserve generation capacity within CLP — taking into account the available transfer capability of the interconnector — in ensuring that HEC’s secondary reserve capacity requirements would be met. The secondary criteria for the two utilities adopted in this Study are therefore almost identical, the only difference being the addition of spinning reserve to the two largest units in CLP’s case. The two secondary criteria described here were adopted for the purposes of this feasibility Study with the aim of providing a consistent analysis between interconnection scenarios, while bearing in mind the historical approach to system planning. They should not be interpreted as the product of a review appropriate to address directly the question of whether, when and how generation support from Southern China should be explicitly included in CLP’s planning process. Given the pace of electricity sector development in Southern China, it is likely that periodic review of this question may be advisable.


26

of secondary planning criteria, nor as an endorsement or criticism of existing criteria. An independent review of the operational criteria may be appropriate to consider the most suitable secondary criteria for the Hong Kong utilities and whether the criteria used by the utilities should be fully harmonised, particularly if a Joint Planning approach is eventually adopted in Hong Kong.

2.4 Technical Feasibility of Increased Interconnection

General Considerations

2.4.1 Section 2.2 outlined many of the benefits that interconnected power systems can achieve, resulting in cost savings for the parties concerned. World-wide, interconnection between electricity utilities is a growing trend. There are few modern power systems currently that are not interconnected with other compatible systems nearby. In fact, Hong Kong is already interconnected with China and Hong Kong Island is currently interconnected to Kowloon. The connection points on Hong Kong and Kowloon are at North Point and Tai Wan respectively.

2.4.2 Whilst some of the above benefits have already been achieved in Hong Kong’s case through the existing interconnector, an assessment of the technical feasibility, system planning and competition issues is carried out in this Study to evaluate the potential benefits of increased interconnection between CLP and HEC that consumers can gain. Before addressing the technical issues, it is worth noting that technical system planning criteria must be applied to ensure that the system will have reliable and adequate supplies of electricity. This is crucial for Hong Kong’s situation given the requirements of the high-rise urban environment and the dependence of key market sectors on a secure and stable electricity supply by the two utilities, CLP and HEC. In the longer term, interconnection also may facilitate the introduction of competition by providing suppliers other than CLP and HEC with access to customers in CLP’s and HEC’s current service areas (see discussion in Section 3).

2.4.3 An interconnection must be capable of technically reliable operation for the potential economic benefits to be realised in practice. There have been a number of interconnection de-coupling incidents in Hong Kong. These are listed in Table 2.4.a. The incidents associated with oscillations or system swings have been a focus of interest in this Study. CLP experienced two such incidents with the CLP-GPHC interconnection before it was upgraded to its existing large capacity in 1992. HEC has experienced one such significant incident. The topic of low frequency oscillations is discussed below in greater detail.


27

Table 2.4.a Interconnector Decoupling Incidents in Hong Kong

Date Cause19 Aug 1983

CLP 2 x 350 MW units disconnected due to tripping of CLP 400 kV double overhead line circuits caused by lightning strike

4 Mar 1984 CLP blackout due to swing (oscillations) in GPHC26 Nov 1987

System swing (oscillations) due to transmission fault in GPHC system

24 Aug 1994

A switch fault at CLP system control centre

19 April 1996

HEC 132 kV overhead line tripped due to lightning strike and activating HEC 3-Phase fault detection scheme

1 May 1996 System swing (oscillations) when HEC’s power system stabilisers (PSS) were taken out of service for testing purposes during the commissioning phase of a new generation unit (1)

Source: HEC and CLP

2.4.4 The technical feasibility of interconnection mainly involves an assessment of the following: load flow, and short circuit analysis; system stability; Firm Capacity and Firm Power; and system reliability.(2)

Load Flow and Short Circuit Analysis

2.4.5 Load flow issues involve an analysis of transmission system loading and voltage control. In assessing load flow, ie the routing of power across an electricity grid, it is critical to ensure that each route used has adequate capacity to carry the quantity of power being transferred to avoid overloading particular electrical circuits. Short circuit analysis is related to fault level study with verification of, for example, switch-gear and circuit breaker ratings.

2.4.6 In assessing the technical feasibility of the current interconnection in Hong Kong and any future additions, load flow analysis has therefore been carried out to: determine the transfer capability of the existing interconnection; identify system enhancements and additional interconnection

plans to increase the existing transfer capability between the two utilities; and

1(?) The utilities have subsequently agreed not to take all PSS units out of service during a system test. 2(?) Level I (generation) reliability has been considered in this feasibility Study. A more detailed Level I and II (generation and transmission) reliability analysis should be conducted subsequently.


28

determine the transfer capability between the two utilities associated with additional interconnection.

2.4.7 In addition, using stability and short circuit analysis, any detrimental impacts on the interconnected systems that may be caused by the system enhancements or the additional interconnection can be identified and the interconnection plans can be modified to eliminate the impact, if any.

System Stability

2.4.8 System stability analysis is used to examine whether a power system will remain stable under a variety of adverse events. It involves consideration of both transient and dynamic stability. Transient stability analysis looks at the effect on the system of large disturbances (loss of lines, generator faults, etc) and requires a complete non-linear electromechanical model of the system to investigate. Dynamic stability analysis looks at the effect of small variations around an equilibrium operating point, and can be investigated using a linear approximation to the full system model. This Study examined basic transient stability with respect to the interconnection of the HEC and CLP systems.

Dynamic Stability — Low Frequency Inter-Area Oscillations

2.4.9 During the course of the Study, a concern was expressed that inter-area low frequency oscillations may present a barrier to increased use of HEC-CLP interconnection. In this context, the Consultants considered whether — in addition to potential economic benefits — greater interconnection capacity between the HEC and CLP systems would mitigate the inter-area low frequency oscillations that have been observed on the system.(1)

2.4.10 Discussions regarding the issues of low frequency oscillations and system stability were conducted with the two utilities. These discussions indicated that low frequency oscillations previously experienced by CLP that led to the de-coupling of their interconnection with GPHC, as documented in Table 2.4.a, occurred at a time when the interconnection capacity between the two systems was small. Since the CLP-GPHC interconnection capacity was upgraded, no such problems have been reported. Analysis therefore focused on the oscillations experienced by HEC.

2.4.11 During a visit to HEC’s Ap Lei Chau control centre, the Consultants observed a continuous oscillation of about 0.7 Hz from the monitoring point on the interconnector, as reported by HEC. Under general operating circumstances, HEC’s installed PSS equipment continually compensates for such oscillations.(2) This oscillation 1(?) These oscillations have been attributed by HEC to inter-area oscillations between the HEC system and the Southern China system.2(?) HEC have studied this continuous oscillation and have concluded that no further refinement can be made to the tuning and operation of the automatic voltage regulation (AVR) and PSS


29

occurs on a magnitude of around 10 MW during normal operation, but has been observed to range up to 1 000 MW in case of system faults.(1)

2.4.12 An eigenvalue analysis was performed to assess the natural damping(2) of the HEC-CLP system as it is currently configured and interconnected with the GPHC system. The analysis used data supplied by the two utilities describing the HEC and CLP networks and generator characteristics(3) and an approximated representation of the interconnected Southern China system.(4) The analysis found that the oscillation mode with the lowest damping (with a damping constant of 0.190 and a time period 1.265) occurs between CLP/GPHC and HEC, with CLP and GPHC in phase.(5) The eigenvalue

devices to overcome it. 1(?) It should be noted that a situation where the PSS is out of service is considered to be a system fault. In the case of the 1 May 1996 incident described in Table 2.4.a, such a system fault was introduced by HEC for testing purposes during the commissioning phase of a new generation unit.2(?) In order to provide sufficient stability margin in case of system faults, structural network changes, unforeseen loading conditions as well as malfunctioning of stabilisation devices, electrical power systems must be designed with sufficient natural damping margin. The natural system damping is basically determined by the system topology, network and generator impedances, generator parameters such as inertia, reactances and time constants as well as system loading conditions. Even if the natural damping is in general positive, superimposed effects of high speed AVRs and power control systems can heavily decrease the resulting damping, making additional measures necessary to compensate for the lag of damping. Most commonly used are Power System Stabilisers (PSS) acting on the VAR set-point by superimposing appropriate composed signals. (PSS have been extensively implemented in Hong Kong by both HEC and CLP. The extent to which generators in Southern China have PSS is not clear, but it is thought that many do not.) However, since PSS devices only act indirectly via voltage dependency of loads — making their efficiency sensitive to system topology and changes in loading conditions — and in order to cope with PSS malfunctioning, it is common practice to design transmission systems with sufficient natural damping margin.3(?) A complete, full-scale, non-linear generator model was used, which accounts for all contributions to damping using 5th- and 6th-order differential equations. This provides a strictly accurate (not approximate) mathematical representation of the generators in a system. This is to be contrasted with the alternative approach that uses a reduced-order representation of generators, which mathematically eliminates damping, thus requiring the inclusion of a so-called “damping factor” approximation, which would not result in a correct determination of the damping.4(?) The modelling result obtained represented quite closely the observed oscillatory behaviour of the system. Sensitivities were conducted at the two extreme ends of the possible range for an approximated representation of the interconnected Southern China system: one with Southern China set to zero (neglected) and one with it modelled as of infinite size. These both produced results further away from the observed system behaviour. It would be possible to conduct a full-scope eigenvalue analysis with a complete (rather than approximated) representation of the Southern China system if the necessary data on the generators, the network and loading in Southern China could be established. However, it is accepted that such data may take some time to assemble. The oscillation modes found in the analysis conducted for this Study would form a sub-set of the results from such a study — a number of other oscillation modes would also be found due to the inclusion of more generators in the analysis. 5(?) Some exploratory eigenvalue sensitivity analysis around this point showed that the alternative oscillation mode where HEC and GPHC each oscillate against CLP is also possible. Although it is considered more probable that HEC will oscillate against CLP and GPHC in phase with one another, the two possible oscillation modes are actually quite close to one another, and so the mode that actually occurs at a particular time depends on how the system is excited (the particular event that initiates the oscillation), and the relative operating state (loading, generator outputs and so on) in each system. It was the alternative oscillation mode that was observed in the 1 May 1996 episode listed in Table 2.4.a, possibly as a result of the particular operating conditions introduced for testing purposes on that occasion. The particular mode of oscillation is not critical, since oscillations would be observed on the HEC-CLP interconnection in both possible oscillation modes.


30

analysis indicates that the observed low frequency oscillation is primarily based on the existing limited capacity, high impedance interconnector between CLP and HEC. Increasing the interconnection capacity between CLP and HEC offers a potential solution, which was analysed as discussed below, although the findings would need to be confirmed by more detailed study.

2.4.13 When oscillations are damped, the amplitude declines from the initial disturbance in an exponential decay function. This decay is rapid for heavily damped oscillations and gradual for lightly damped oscillations. Undamped oscillations exhibit exponential growth in amplitude. The degree of damping is conveniently measured by comparing the ratio of the amplitudes of two subsequent peaks in the oscillation. The damping constant and the time period of the oscillation mode calculated as described above result in a peak/peak amplitude ratio of 1.272.

2.4.14 While there is no hard and fast line dividing adequate from inadequate damping, the initial eigenvalue analysis indicates that the current system may be considered marginally stable with the present level of damping. If all other factors in a power system, including network configuration, are held constant, then each addition of generation reduces the damping slightly. Alternatively, if all other factors, including generation, are held constant, then increasing interconnection capacity tends to increase damping. Consistent with these general principles of power systems, further eigenvalue analysis showed that increased interconnection capacity will eventually be needed to increase the damping as generation units are added to the system, although this would not necessarily be required before the next individual unit addition.

2.4.15 Consequently, upgrading of the interconnector would serve several purposes simultaneously, allowing: increasing flexibility in generation expansion planning including

siting future power stations, thereby opening opportunities for the implementation of lower cost generation/transmission expansion from an integrated system point of view (under Joint Planning);

more economical system operation (under Joint Dispatch); an increase in the overall Level I and Level II system reliability;

and increased natural system damping, allowing the interconnected

system to overcome the existing low-frequency inter-area oscillation problem.

Upgrade of the HEC-CLP interconnection is assessed from an economic point of view in sub-section 2.5.

2.4.16 Alternative technical options to overcome low frequency oscillations, such as a direct current (DC) interconnector have been suggested. A DC interconnector is considered technically feasible and would potentially eliminate the issue of low frequency AC oscillations.


31

However, on short interconnections, the AC-to-DC and DC-to-AC inverters incur high energy losses compared with AC interconnectors. DC interconnections are also costly and unlikely to be economic in this instance. Costs estimates for a DC interconnector would range upwards from $2.0 billion. Since the eigenvalue analysis showed that a standard AC interconnection with increased capacity would increase natural damping to the extent that low frequency inter-area oscillations are overcome, a DC interconnection is clearly not the optimum economic solution to the problem.

2.4.17 During the course of the Study, it was suggested to the Consultants that a rule of thumb known as the “Concordia Rule” should be used as a basis for calculating the size of an increased interconnector. This ‘rule’ states that the size of an interconnector between two systems should be five times the size of the largest unit in the smallest system. In the case of the present interconnected Hong Kong system, for example, the largest generation unit in the smallest system is a 350 MW HEC unit. Based on such a calculation, Hong Kong’s HEC-CLP interconnection would require a capacity of 1 750 MW (or five times 350 MW) for Firm Power. Examples of interconnection capacity from developed country power systems were provided to the Consultants that meet the Concordia Rule, although this is most likely circumstantial. It is the Consultant’s view that, since the Concordia Rule is an empirical relationship observed after the fact, it has not been actually applied to determine the capacity of interconnection in the various examples. The Consultant is of the opinion that the Concordia Rule has no theoretical basis, except that it acknowledges that an increased size of interconnector would increase natural system damping.

South China and Hong Kong Electricity Reliability Council

2.4.18 During the Study, consideration was given to the need for an electricity reliability co-ordinating council to be established for the Southern China region, including Hong Kong. A reliability council similar to those in other countries would help the Southern China region achieve a higher level of reliability and system operation transparency, a critical technical step in the process towards competition,(1) as discussed in Section 3. Such a council could provide benefits through the following: establishment of technical standards for generation and

interconnection planning; definition of terms of measurement or system performance

necessary to meet the standards; guides that describe good planning practices for consideration by

council members;

1(?) The authorities in Mainland China have announced plans to restructure its electricity supply system and to separate the ownership of generation from transmission, and from government ownership. These issues are discussed further in Section 4.


32

specifying how to comply with the standards; and addressing issues regarding implementation, compliance and

enforcement of the standards.

Background to Hong Kong Interconnection

2.4.19 The interconnection between Hong Kong Island and Kowloon linking the CLP and HEC systems has been in place since April 1981. This interconnector allows emergency support, economy power interchanges, sharing of spinning reserve, and mutual backup of both systems. At present, the interconnection comprises six 132 kV, 120 MVA submarine cable circuits with a total rated capacity of 720 MVA arranged in three circuits across the harbour between CLP’s Tai Wan 400 kV Substation and HEC’s North Point 132 kV Substation. The submarine cables themselves are owned jointly by HEC and CLP, with all associated land, equipment and cables owned by the companies in whose geographic area they reside.

2.4.20 CLP’s 400 kV network is further connected to the Daya Bay Nuclear Power Station and to the Guangzhou Pumped Storage Plant via the 500 kV network in Guangdong Province. Since 1992, this link has been used as the normal interconnection path between CLP and Guangdong Electric Power Holding Company (GPHC) in Mainland China. This interconnection is used to deliver CLP’s share of the nuclear and pumped-storage energy from the Mainland to the CLP system and for bulk energy interchanges from CLP to GPHC. CLP and GPHC are also interconnected at 132/110 kV level as back-up circuits to supply power from CLP to GPHC. The total MVA of the interconnection between CLP and GPHC is almost 5 000 MVA.

2.4.21 The HEC-CLP interconnector is small (rated capacity 720 MVA) when compared with HEC’s peak demand of around 2 200 MW and installed capacity of around 3 300 MW. This has a bearing on the admission of new power generation utilities, should the Hong Kong wholesale electricity supply market become open to competition in the future, given that new players in a future competitive wholesale electricity market in Hong Kong are likely to be located in Mainland China. This is due to the fact that the interconnection capacity would constrain the amount of the load south of the harbour that CLP and any new competitive market entrants could serve.(1) CLP’s link to the Mainland (almost 5 000 MVA) would not be similarly constrained.

2.4.22 The operation and maintenance of the HEC-CLP interconnector and its metering, service conditions, billing and payment is done through a mutual agreement established between the two utilities known as the Interconnection Agreement. This covers issues such as ownership regulation, power exchange services (emergency exchange, economy exchange and firm exchange) and installed

1(?) Under the reasonable assumption that competitors would be unable to build new power plants inside HEC’s transmission network.


33

generating capacity. The Interconnection Agreement is a voluntary arrangement that has been reached between the two utilities after some initial involvement by Government. Government is not a party to the Interconnection Agreement, but has some continuing, but limited oversight of this arrangement indirectly through the SCAs.

Capability of Existing HEC-CLP Interconnection

2.4.23 The original rating of Hong Kong’s interconnector when installed in 1981 was two circuits of 240 MVA. The third cable was added later, bringing a total rated capacity of 720 MVA.(1)

2.4.24 Allowing for a single (or N-1) “contingency” — the loss of one 240 MVA circuit — allows the calculation of the Total Transfer Capacity (TTC) under N-1 conditions. Using a power factor of 0.9 to be on the conservative side, this would result in a TTC of (3-1)*240 MVA * 0.9 pf = 432 MW.

2.4.25 However, based on an assessment of load flow, the system characteristics and constraints, Available Transfer Capability (ATC) differs from the TTC as follows: 217 MW Available Transfer Capability from CLP to HEC.

Transfer capability limitations are caused by the assumed contingency loss of one of the fully loaded Lamma 350 MW units. This loss would immediately be met by spinning reserve, which is carried on a shared basis of 135 MW at HEC and 240 MW at CLP. Therefore, under the worst generation contingency just discussed, HEC would require spinning reserve support from CLP of 350 - 135 = 215 MW. Therefore, to ensure that the CLP to HEC spinning reserve obligation can be met at all times, 215 MW of capacity on the interconnector would be ‘set aside’ for spinning reserve support. The ATC is therefore 432 - 215 = 217 MW.(2)

297 MW Available Transfer Capability from HEC to CLP. Available Transfer Capability is limited by thermal overloading of the two circuits from Tai Wan to North Point B bus 132 kV circuit, for the loss of the Tai Wan to North Point A 132 kV circuit and the consideration of a 135 MW contribution from HEC to CLP in case of a generation deficit at CLP.(3)

1(?) One MVA at a power factor of 1 is equal to 1 MW. The power factor across the interconnector is usually close to 1.2(?) This is a very conservative calculation of the available transfer capability of the interconnector for two reasons. Firstly, it allows for the contingency loss of one interconnector circuit and the simultaneous contingency loss of a generation unit. This is known as a double contingency — in this case a combined generation and transmission N-2 contingency. Secondly, the interconnector cable rating used in the calculation is the “normal” rating, which describes the ability of the cable to deliver load over an indefinite period. Cables can safely be loaded up to 10% higher than their normal rating for short (up to four hour) periods. This is known as the “cyclic” rating, as such higher loading can be applied cyclically. It should be noted that when cyclic loading conditions are applied, the ATC is 260 MW at power factor 0.9, and up to 313 MW if power factors of up to 1.0 are considered. 3(?) Again, if “cyclic loading” and probable power factors of 0.9 to 1 are considered, ATC can be viewed to be 340 MW up to 393 MW.


34

These limitations on transfer capability are taken into consideration when calculating the amount of reserve capacity required to meet the secondary criteria of each utility.(1)

2.4.26 In 1995, CLP provided up to 485 MW(2) for four hours to HEC through the interconnector as they had lost output from one coal-fired unit at the Lamma Power Station. HEC have explained that this was an economic interchange as support across the interconnector was not necessary on that occasion, but that it was cheaper than the alternative of running gas turbines at HEC. However, the example shows that the actual amount of power that can be transferred in practice is higher than the calculated transfer capability, but not higher than the rated capacity. This is because the transfer capability is calculated assuming the contingency loss of one circuit of 240 MVA, which is standard power system planning practice internationally.

Benefits Realised with the Existing Interconnection and Additional Benefits Potentially Available

2.4.27 Various categories of benefits are potentially available under Firm Capacity and Firm Power. They include: reduced reserve margin requirements with reliability maintained; reduced spinning reserve requirements with reliability

maintained; provision of emergency support (short-term disruptions); security of supply (Long-Term Disruption); and fuel costs savings.Each of these categories of benefits are discussed below, along with comments on whether those benefits are currently being realised in Hong Kong and an indication of the significance of additional benefits that are potentially available in Hong Kong.

Maintaining Reliability while Reducing Reserve Margins

2.4.28 To ensure that sufficient capacity is available to meet electricity demand, electric utilities need to install more generating capacity than would be required to just meet their customers’ peak load. This reserve generating capacity is often referred to in percentage terms as reserve margin, which is the percentage by which the

1(?) At a late stage of this feasibility Study, it was suggested that the mutual heating effect of underground cables on the land side of the existing interconnector circuits is such that the capacity of the interconnector should be de-rated. The results of investigation of the mutual heating effect are not considered conclusive at this stage, and Government has asked the concerned utility for further information and a plan of action on this question. Due to the inconclusive nature of the information available at the time of this Study and the Consultants’ view that it is not appropriate to consider mutual heating effects in one network location and not in other locations, the transfer capabilities used in this feasibility Study have not de-rated the existing interconnector to allow for possible mutual heating effects. 2(?) The link could be run at a higher capacity than its rated capacity for a short period only.


35

installed capacity in a given year exceeds the expected maximum demand in that year. Reserve capacity is provided for two main reasons: to meet electricity demand during planned outages (when plant is shut down for maintenance) and also during unplanned or forced outages when unexpected faults cause a generator to “trip.” The degree of flexibility in scheduling generating unit maintenance depends on the amount of reserve plant available. Calculating the required level of reserve generating capacity when planning generation requirements takes into account both of these considerations. Optimum maintenance schedules are found taking into account the times of year when demand is low. Forced outages are taken into account by LOLP analysis.

2.4.29 Interconnected utility systems can share reserve plant capacity in such a way that they both meet their minimum reserve capacity requirements, calculated taking into account both planned and forced outage, with a lower overall level of reserve plant capacity in each individual utility than would be required if they each provided sufficient reserve plant independently. The reserve plant would then represent a smaller proportion of the peak load, so the reserve margins required for a given level of reliability would be reduced by the interconnection.

2.4.30 The existing interconnector has already resulted in lower reserve margins. However, the potential exists for further reductions in this area from increased interconnection capacity.

Reduced Spinning Reserve

2.4.31 Spinning reserve is generating capacity that is kept in operation although it is not needed to meet expected demand. Its function is to replace immediately the generation lost from any plant or transmission line that unexpectedly trips (an unplanned or “forced” outage event). The amount of spinning reserve (which is generally less than the built-in reserve margin) necessary to assure reliability is primarily a function of the size of the largest unit in a system because that is the greatest amount of generation that can be lost in one outage.

2.4.32 Carrying spinning reserve is costly to utilities because it requires operating some plants at load points where their efficiencies are not maximised. Therefore, reducing the required amount of spinning reserve results in cost savings by decreasing the number of generating units that would otherwise be run at inefficient levels.

2.4.33 Through the installation of the interconnector between the two utilities, CLP and HEC have reduced the amount of spinning reserve they must carry, as the two systems can be treated as one for reserve purposes. In the absence of the interconnector, each company would have to carry its own full level of spinning reserve equal to the largest unit on their system. HEC would therefore have to provide for the loss of one of its 350 MW units and CLP would


36

correspondingly have to provide for the loss of one of its 680 MW coal units. As a result of CLP’s interconnection to the Guangdong power system, spinning reserve allowance must be made for loss of output from one of CLP’s 985 MW nuclear units at Daya Bay, as the largest operating unit on the interconnected HEC-CLP-GPHC system. Therefore, in combination there are three utilities — CLP, HEC and GPHC — who can combine to carry spinning reserve on a shared basis..

2.4.34 In addition to simple sharing of spinning reserve, consideration of the frequency-dependence of load in calculation of spinning reserve requirements allows for further benefits of interconnection. This approach, adopted by HEC and CLP in their interconnected system with GPHC, may be explained as follows: Frequency must always be held constant over an interconnected

electricity system. Therefore, when a generator trips due to a fault somewhere in the

system, the frequency of the entire interconnected system will reduce automatically in response.

Load is dependent on frequency. The load-frequency dependence is typically greater than or equal

to 1% per % (a 1% load reduction for every 1% frequency reduction). In other words, if the system frequency was reduced by 1% of 50 Hz, which is 0.5 Hz, then the load on the entire interconnected system would automatically be reduced by 1%.

The frequency-dependence varies according to the type of load — it is low for loads such as lights and high (around 3% per %) for motor loads. The different proportions of each type of load in different areas therefore means that the frequency-dependence of load is slightly different in HEC, CLP and GPHC. It is around 1.5% per % for CLP and closer to 1% per % in the HEC system.

In the under-frequency load shedding system presently in place, the first stage involves shedding 5% of load when the frequency on the interconnected system has fallen to 48.8 Hz, which is 2.4% below the standard frequency of 50 Hz.

Assuming a system-wide weighted average load-frequency dependence of 1.5% per %, frequency-dependent load reduction will be therefore be 3.6% of the load at that time.

The worst case — that with the lowest frequency-dependent load reduction — occurs during the time of lowest load. The lowest load in the HEC-CLP-GPHC system occurs at night during Chinese New Year when many people are away on vacation, loads are low anyway during winter and loads are lowest at night. At this time the HEC-CLP-GPHC load is approximately 6 800 MW(1). The worst-case frequency-dependent load reduction will therefore be

1(?) This is made up of about 500, 1300 and 5 000 MW for each of HEC, CLP and GPHC respectively.


37

3.6% of 6 800 MW, or about 245 MW. This represents a spinning reserve saving.

The worst-case single generator contingency is the trip of one of the 985 MW Daya Bay nuclear units. In this event, the frequency-dependent load reduction would be greater than or equal to 245 MW. Therefore, the maximum spinning reserve that must be carried at any one time is 740 MW.(1)

In this worst-case, the system would operate for some minutes marginally above the limit for the first stage of under-frequency load shedding. Mobilisation of ”slow” power increase (as distinct from instantaneous spinning reserve) would then recover the system frequency to its nominal value.(2)

2.4.35 The current Hong Kong interconnector is capable of providing full spinning reserve benefits and hence further benefits in this area from increasing the interconnector capacity are unlikely. See Table 2.4.b for a summary of the reductions achieved in spinning reserve from both the increase in frequency-dependent load reduction due to HEC-CLP-GPHC interconnection and those made possible by sharing spinning reserve between HEC, CLP and GPHC.

Table 2.4.b Reduction Achieved in Spinning Reserve Due to Existing Interconnector

System Largest Unit On Individual

System

- Maximum Spinning Reserve

Carried(3)

= Spinning Reserve Saved(4)

HEC-CLP-GPHC 985 MW 740 MW 245 MW

HEC (5) 350 MW 135 MW 215 MW

CLP 680 MW 240 MW 440 MW

GPHC 985 MW 365 MW 620 MW

1(?) At times of higher load, the frequency-dependent load reduction would be greater, so the total amount of shared spinning reserve that would need to be carried at those times would also be lower.2(?) The generation deficit would be covered by fast starting units like gas turbines and pumped storage, increasing boiler loading in the case of sliding pressure control and using hot standby3(?) The individual utilities’ spinning reserve contributions shown in this column are additive. 4(?) The spinning reserve savings shown here in each of the individual rows are not additive. The spinning reserve savings for the interconnected system as a whole due to consideration of the system-wide frequency-dependence of load are shown at the top of the table. The values for the individual utilities include both the savings from sharing and the savings due to the frequency-dependence of load benefits that flow from having an interconnected HEC-CLP-GPHC system. None of these benefits would be possible if the three utilities were not interconnected. 5(?) This value will increase to 375 MW after the currently planned re-powering of two coal units at Lamma power station in 2001 and 2002. HEC will still need to carry just 135 MW of spinning reserve, so this will increase the spinning reserve savings experienced by HEC relative to the spinning reserve that they would need to carry if they were an isolated system by 25 MW to 240 MW.


38

Emergency Support

2.4.36 Hong Kong’s interconnector provides emergency support between the two utility power systems. In 1995 HEC purchased a significant amount of electricity during a problem with one of its boilers. Although HEC have explained that their gas turbines could have been run to cover the load on that occasion, this event is an example of how the emergency support role would allow one utility’s power system to be able to obtain support from the other’s to meet demand under unforeseen circumstances. Considering Hong Kong’s high urban population density in a humid tropical climate with high-rise commercial and residential buildings dependent on electricity for lifts and air conditioning systems, as well as Hong Kong’s position as one of the world’s leading financial and commercial centres, reliability of electricity supply is clearly paramount for this society. The Consultants consider that significant emergency support benefits available from interconnection are already being realised with the existing interconnector. However, these reliability benefits could certainly be extended by the installation of an additional, separate interconnector at a higher voltage, based on Level II (transmission) reliability considerations as discussed under Transmission in the Methodology and Approach sub-section above.

Generation Plant Economies of Scale

2.4.37 Smaller power systems tend to rely on smaller power stations. If large generating units are introduced into small systems then the situation arises where such units would be operating inefficiently at part load for much of the time, because demand is not great enough to call upon them for long periods. This results in higher reserve margins than necessary and, as demand grows, the step increases in generating capacity do not match demand increases. This means that smaller systems are constrained to use smaller generating units, which generally have higher operating costs per unit of electricity produced.

2.4.38 Generating units reach an optimum economic size between 300 MW and 700 MW, eg these size units allow the lowest fuel and capital costs to be attained. When power systems can utilise generating units of these sizes then there are no further economies of scale to exploit as the individual power system grows due to demand increases. Hence, interconnectors which join together two large systems generally do not lead to further economies of scale in power station design. However, if linking two small systems or one large and one small system, interconnection does allow the advantage of economies of scale in the installation of larger size generating units than would otherwise be possible.

Provision of Emergency Support (Short-Term Disruptions)

2.4.39 Even the best designed power systems cannot avoid failures for various reasons. In the event of a short-term disruption such as an ERM GOVERNMENT OF THE HKSAR

39

unplanned or forced outage caused by failure or “tripping” of a generating unit, transmission line or interconnector, utilities generally design power systems to include a margin of reserve capacity over expected peak demand to compensate for this loss. This reserve capacity is called into operation to provide Emergency Support or Backup.

2.4.40 Systems are planned using probability analysis to allow for forced outages. , Basic probability theory shows that the probability of two events occurring simultaneously is much lower than the probability of either event occurring individually. Hence, although it is possible, it is very unlikely that two systems will have significant simultaneous failures and have to call upon their backup generating units at the same time. Therefore, if the two systems are interconnected, the utility with the affected system can call upon the reserves of the other utility/ies to provide temporary support. The overall system need for reserve margins is thereby reduced. These considerations are taken into account in the probability analysis used in generation planning.

Security of Supply (Long-Term Disruption)

2.4.41 In the long term, supply of electricity can sometimes be affected by unforeseen changes that may have major implications for a utility’s power system security, such as significant escalation in fuel prices, unavailability of fuels (eg, during embargoes, civil strife or natural catastrophes, etc).

2.4.42 An interconnected system of two or more utilities can help assure the security of supply in case of such disruptions. This is particularly the case when two such systems in combination have a mix of fuel supplies. For example, the situation could arise where one utility is highly dependent on coal while the other is dependent on natural gas. If there is a disruption of supply of coal then, with an interconnected system, the gas-based system can support the coal-based system. The reverse would be true if the gas supply is disrupted.

2.4.43 Security of supply issues are nominally related to the disruption of a particular supply of fuel due to events such as embargoes, wars, natural disasters etc as, for example, during the oil crisis of the early 1970s. Since both utilities rely heavily on coal for their current fuel needs, the full benefits of diversity may not be fully achieved. At present, CLP has a broader diversity of fuel types than HEC. In addition to coal it has natural gas (at Black Point) and nuclear (at Daya Bay) generation plant. These would continue to provide supply in the event of disruption to coal supplies. HEC will achieve further benefits of diversity if it implements a scheme to use liquefied natural gas (LNG) at its proposed Lamma Extension. The use of LNG for the proposed extension has been proposed with Government’s encouragement, mainly to address environmental concerns. Emissions to air from natural gas-fired electricity


40

generation are lower than from coal-fired generation, which is significant, given the increasing level of concern about the environment and air quality in particular.

2.4.44 Quantifying in economic terms, the benefits due to interconnection of security of supply and fuel diversity — including environmental benefits — is a substantial undertaking, however. They need to be factored into any major decisions regarding changes to the existing regulatory system. Access to the more diverse supplies of fuel from Mainland China through interconnection could be a major benefit, if the appropriate institutional arrangements and environmental controls are in place, should security of fuel supply become a key issue.

Fuel Cost Savings

2.4.45 During Firm Power transfer between interconnected systems, fuel cost savings could accrue for several reasons. For instance, savings would arise if two utilities with different peak demand times could be linked so that a flatter load profile could be derived in a combined system. This means that base-load generating units from either system, which are generally more efficient and have lower fuel and capital costs per kWh produced, can be better utilised. An example of this is if one system is dominated by residential consumption, and which experiences power peaks at 7 pm, is combined with another system dominated by commercial consumers with peak characteristics at 11 am. The combined system would have a much flatter load profile through the course of the day and hence lower fuel costs could be shared between the utilities concerned.

2.4.46 An interconnected system can further take advantage of differences in fuel costs between two systems. If, for example, one utility has better access to less expensive natural gas — and thus can generate electricity more cheaply — than the other, then the interconnector can be used to transmit the cheaper electricity to the other system. This becomes apparent if arrangements are made concerning the short term trading of economy power and the long term purchasing or selling of power during periods of capacity deficiency or surplus between the interconnected utilities. It is also dependent on the types of contracts that the utilities have signed to purchase the fuels, particularly in the case of gas and the conditions specified in their agreement with the regulatory authority.

2.4.47 The interconnector has provided limited fuel cost savings to date through economy power exchange. Economy power exchange is the trading of power that often takes place where one utility avoids starting up a more expensive generating unit if another has a less expensive one either running or ready to run, eg Firm Power on a limited scale. The savings that have been achieved to date are a very small proportion of the actual amounts spent on fuel. For instance, HEC and CLP combined spent approximately $3.3 billion on


41

fossil fuels in 1996 and fuel savings were only $1.4 million: less than0.1% of the fuel expenditure. The opportunities for further savings are limited by a number of constraints in Hong Kong on economy exchange of electricity between HEC and CLP. These are discussed in paragraph 2.5.9 below in the context of limitations on potential economic benefits from Joint Dispatch. shows the savings arising from fuel costs savings through economy exchange for the period 1990 to 1996.

2.4.48 The savings that have been achieved to date are a very small proportion of the actual amounts spent on fuel. For instance, HEC and CLP combined spent approximately $3.3 billion on fossil fuels in 1996 and fuel savings were only $1.4 million: less than 0.1% of the fuel expenditure. The opportunities for further savings are limited by a number of constraints in Hong Kong on economy exchange of electricity between HEC and CLP. These are discussed in paragraph 2.5.9 below in the context of limitations on potential economic benefits from Joint Dispatch.

Table 2.4.c Historical Fuel Cost Savings from Economy Exchanges

Year HEC CLP Total$ million $ million $ million

1990 2.0 1.6 3.61991 4.9 5.3 10.21992 6.6 5.7 12.31993 5.5 6.3 11.91994 1.5 1.4 2.91995 2.7 3.1 5.81996 0.7 0.7 1.41997(1) 0.0 0.0 0.01998 0.0 0.1 0.1

Source: Economic Services Bureau

Greater Scope for Competition

2.4.49 Consideration of whether or not the existing interconnection capacity would be a constraint to competition and whether or nor it would be necessary to increase the interconnection capacity to provide for successful operation of a possible future competitive electricity market is discussed in Section 3 on the feasibility of introducing electricity sector competition in Hong Kong.

Summary of Additional Benefits Potentially Available from Interconnection

2.4.50 A summary of additional benefits potentially available from interconnection is provided in Table 2.4.d.

1(?) HEC and CLP did achieve some fuel cost savings from economy power exchanges in 1997 and HEC did in 1998, but since these were less than $50 000 they round off to 0.0 million.


42

Table 2.4.dSummary of Additional Benefits Potentially Available from Increased Interconnection Capacity

Use of inter-connection

Benefit Description of Benefit

Firm Capacity Reduced reserve margin

Increased interconnection capacity allows the reliability criteria to be met with less reserve plant capacity by co-ordinated or joint capacity expansion planning.

Emergency support Additional benefits achieved mainly through optimised transfer of reduced overall reserve margins and, for each individual utility, increased diversification of locations of supporting plants and transmission supply paths.

Firm Power Security of supply Strengthening would provide further assurance through access by HEC to more diverse fuel supplies from CLP and the Mainland, and CLP’s access to HEC’s LNG supplies, should HEC acquire a supply of LNG.

Fuel cost savings Further savings available through Firm Power.

Economies of scale Would be feasible with both Firm Capacity and Firm Power.

Enabling greater scope for competition

If electricity sector competition was to be introduced in Hong Kong, the existing interconnection may constrain the competitive operation of the market. An increase in interconnection capacity would reduce such a constraint and could therefore allow incremental benefits from electricity sector competition to be realised. Such potential future benefits are not quantified in this Study, but the broad range of issues with respect to the feasibility of introducing electricity sector competition in Hong Kong are discussed in Section 3.

Technical Options for Upgrading the HEC-CLP Interconnection Capacity

2.4.51 A number of benefits have been achieved through the existing interconnection between Hong Kong Island and Kowloon. However, due to the limitations of this interconnection, potential for Firm Power is limited and the main advantages to be gained through Firm Power, Firm Capacity and competition have not and cannot be realised. The enhancement of the interconnector through additional


43

capacity would technically enable access between the two utility systems to help achieve these benefits provided the economics are attractive.

2.4.52 As noted under the earlier section discussing the capability of the existing interconnector, the flow on the existing interconnector is not balanced. This limits the transfer capability of the interconnector. The Consultants consider that several alternative technical solutions may be available to balance the flow on the interconnector, which would increase the transfer capability somewhat to approximately 430 MW. Some of these solutions require additional equipment, others would require only operational changes. Rather than present a detailed evaluation of each option, it is conservatively assumed for the purposes of the feasibility study that the most expensive of these options is adopted. This option involves the addition of a series reactor on the existing interconnector. This upgrade would be installed for all interconnection cases considered. The reference interconnection case — the “Base Case” was therefore defined as that with the flow balanced on the existing interconnectors, but otherwise no upgrade to the existing interconnection configuration.

2.4.53 The Study then considered the following options for increasing interconnection: a fourth 240 MVA 132 kV circuit from Tai Wan to a new HEC

substation in the Wan Chai area with series reactors; and two new 550 MVA 275 kV cables separate from the existing

interconnectors, including Phase Angle Regulator (PAR).

2.4.54 Interconnection upgrade cases were defined as: Case A, involving installation of the fourth 132 kV cable in 2004; Case B, involving installation of the two by 275 kV upgrade in

2004; and Case C, which is the same as Case B, but would also upgrade the

132 kV interconnector in 2008, by the addition of a fourth 132 kV cable.

Each of these alternatives, and a combination of them, is discussed below.(1)

2.4.55 The addition of the fourth 132 kV circuit to the existing interconnector would increase the total transfer capability from 432 MW to 648 MW. The available transfer capability from CLP to HEC, with the existing largest generation unit size, would increase from 217 MW to 408 MW after the re-powering of HEC’s 350 MW

1(?) The values given here are calculated using the same method as that described for the calculation of transfer capability for the existing interconnection. Normal loading is used (rather than a 1.1 cyclic loading factor) and the same fairly conservative 0.9 power factor is assumed to convert MVA to MW. Although exact values are given here, interconnection and generation capacities are rounded to the nearest 10 MW in the 2-area LOLP analysis software used in this feasibility Study.


44

units to 375 MW. The available transfer capability from HEC to CLP, would increase from 297 MW to 513 MW.(1)

2.4.56 The addition of two new 275 kV circuits, without the addition of a fourth 132 kV circuit, would increase the total transfer capability from 432 MW to 1 143 MW. The available transfer capability from CLP to HEC, with the existing largest generation unit size, would increase from 217 MW to 903 MW after the re-powering of HEC’s 350 MW units to 375 MW. The available transfer capability from HEC to CLP, would increase from 297 MW to 1 008 MW.(2)

2.4.57 The addition of both two new 275 kV circuits, and a fourth 132 kV circuit, would increase the total transfer capability from 432 MW to 1 359 MW. The available transfer capability from CLP to HEC, with the existing largest generation unit size, would increase from 217 MW to 1 144 MW or 1 119 MW after the re-powering of HEC’s 350 MW units to 375 MW. The available transfer capability from HEC to CLP, would increase from 297 MW to 1 224 MW.(3)

Practical Considerations for Upgrade of Interconnection Capacity

2.4.58 Installing a major infrastructure project such as submarine interconnector cables involves a number of practical and logistical considerations relating to: finding a suitable route corridor for the sub-marine portion of the

interconnector that does not impinge on existing sub-harbour infrastructure corridors such as those of the road and MTR tunnels and submarine water pipes;

finding suitable cable routes for the land portion of the interconnector;

finding suitable landing points for each end of the sub-marine cable;

identifying or obtaining a site and constructing suitable substation facilities at either end of the interconnector;

obtaining the necessary statutory permits from the relevant Government authority/ies;

co-ordinating installation with other infrastructure projects in the area, such as land reclamation; and

1(?) TTC = (4-1)*240MVA * 0.9pf = 648 MW ATC to HEC = 648 - (375-135) = 408 MW ATC to CLP = 648 - 135 = 513 MW 2(?) TTC = [(2-1*550) + 3*240MVA] * 0.9pf = 1 143 MW ATC to HEC = 1 143 - (375 - 135) = 903 MWATC to CLP = 1 143 - 135 = 1 008 MW 3(?) TTC = [(2-1*550) + 4*240MVA] * 0.9pf = 1 359 MW ATC to HEC = 1 359 - (375 - 135) = 1 119 MWATC to CLP = 1 359 - 135 = 1 224 MW


45

managing the logistics of installation, particularly across the busy waterway of Victoria Harbour.

2.4.59 The Consultants are well aware of the complexity of such an exercise, including the planned reclamation in south-east Kowloon that will require the relocation of the existing interconnector cables, and the proposed future reclamation of the Central-Wan Chai area between the outlying island ferry piers and the Convention Centre. However, it was not sensible within the time and scope of this initial feasibility Study to devote significant resources to the detailed minutiæ of the site and route identification and detailed logistical planning that would be required for the construction of additional interconnection capacity.

2.4.60 The Consultants have identified several interconnection options which appear to be feasible at the conceptual level. CLP are of the general view that an additional cross-harbour interconnector could be constructed in a two to three year time frame, provided that there were no undue delays in obtaining the relevant statutory permits from Government. HEC has expressed the view that it would take longer than five years to build the 2 x 550 MVA interconnection circuits at 400/275 kV from Yau Ma Tei to Wan Chai, the delay coming from the Hong Kong side construction work. They have suggested that there is no feasible landing point at the new Central-Wan Chai Reclamation.

2.4.61 The Consultants are aware that consultation is taking place within Government over the timing aspects of the interconnection construction, given the congestion issues on the Hong Kong side. HEC have indicated that although Yau Ma Tei is on newly reclaimed land with no existing infrastructure, the Hong Kong side landing point would have to be on the new Central Reclamation which is bound by the airport railway and is compounded by the need for the cable route to go through the busy Central district before reaching Wan Chai. The Consultants have developed the interconnection cases for analysis of the interconnection/generation scenarios on the assumption that it should be possible to construct an increased interconnector ready for commissioning in time to meet the 2004 peak. However, this would need to be confirmed by a detailed engineering study of the constraints.

Options for System Planning and Operation and use of HEC-CLP Interconnection to Realise Potential Overall Economic Benefits

2.4.62 The potential economic benefits from increased interconnection are dependent on the way that the system is planned and operated. Joint Planning of generation plant can provide economic benefits via the sharing of generation capacity across the interconnection. Such sharing requires that the interconnection support Firm Capacity. This refers to the ability of the interconnection to transfer energy


46

reliably when required (usually for relatively short periods or for a relatively small proportion of the hours in the year) of at least the specified capacity. Joint Dispatch of generation plant can provide economic benefits by enabling the dispatch of the least cost mix of plant for the system as a whole, which by definition provides for production costs less than or equal to the least cost mix in each separate individual area. Such an approach to system operation requires that the interconnection transfer Firm Power. This refers to the ability of the interconnection to transfer energy reliably on a regular basis. It may be possible to obtain economic benefits from both Joint Planning and Joint Dispatch, since Firm Capacity and Firm Power are not mutually exclusive. However, the use of HEC-CLP interconnection to support each is discussed separately below.

Joint Planning — Using Interconnection for Firm Capacity Support

2.4.63 The current interconnector between HEC and CLP is used mainly for Firm Capacity purposes such as emergency support and reduced spinning reserve, two inter-related benefits (see Table 2.2.a). Firm interconnection capacity allows inter-area generation capacity support to be provided should it be needed. Such arrangements may not ever be utilised to transfer power but may be used to allow shared access to reserve capacity, without which the construction of more generating units would be required. The least cost level of generation capacity needed in each area to meet the reliability criteria with a given interconnection capacity is determined by a two-area Loss of Load Probability (LOLP) analysis as described in sub-section 2.3 Methodology and Approach.

Joint Dispatch — Using Interconnection for Firm Power Transfer

2.4.64 The utilities in Hong Kong dispatch their generation plant independently, rather than jointly. Indeed the existing interconnection is considered inadequate for Joint Dispatch of plant due to its relatively small capacity and the associated risk of low frequency inter-area oscillations making it unsuitable for Firm Power transfer. Since there would be more frequent flow of energy under Firm Power transfer (refer Table 2.2.b) —via economy exchange, bilateral contracts or full Joint Dispatch — than when simply relying on an interconnector for emergency generation support under Firm Capacity, the risk of low frequency inter-area oscillation incidents with the existing interconnection would also be higher.(1) Reliance on the existing interconnection for Firm Power transfer is therefore not considered advisable with the existing interconnection capacity. Substantially increasing the interconnection capacity — as in

1(?) HEC have stated that — at least with the existing HEC-CLP interconnection — they might face the possibility of having to shed load (ie cut electricity to customers) if power system instability from low frequency oscillations was to arise while electricity was being transferred across the interconnector under bilateral contracts or Joint Dispatch. An initial eigenvalue analysis of the low frequency inter-area oscillation problem has been conducted as explained in the sub-section 2.4 Technical Feasibility of Increased Interconnection. The results of this analysis confirm HEC’s view that Firm Power transfer is not technically advisable with the existing interconnection capacity.


47

interconnection/generation Scenario B — would help overcome the inter-area low frequency oscillation problem to the extent that use of the interconnection for Firm Power transfer would make Joint Dispatch technically possible.(2)

Combinations of Upgraded HEC-CLP Interconnection Capacity and Operation Modes Studied

2.4.65 An interconnection can be operated or used in such a way that it delivers benefits in terms of overall generation capacity costs (under Joint Planning) or benefits in terms of energy production cost (under Joint Dispatch) or both. This Study examined in detail the benefits potentially available from generation deferral scenarios under several interconnection cases. The mode of operation that would provide benefits of this type is referred to in this Study as Firm Capacity. It should be noted that in this report, this term refers to the use of the interconnection itself for providing available reserve capacity from each utility to the other, rather than generation-level commitments between the utilities: generation capacity support would be provided on a mutual basis using the Firm Capacity of the interconnection based on the probabilistic availability of plant as modelled under LOLP analysis. The mode of operation that would provide production cost benefits is referred to in this Study as Firm Power. This name arises from the use of interconnection for firm transfers of energy.

2(?) The initial eigenvalue analysis mentioned above indicated that this was the case as explained in sub-section 2.4 Technical Feasibility of Increased Interconnection. If Government or either of the utilities consider additional confirmation of the result of the initial eigenvalue analysis necessary to plan for Firm Power before the new interconnector is in operation, then a detailed analysis of the low-frequency oscillation pattern is advised. This task would mainly include the analysis of the oscillation pattern and the utilities’ participation in the oscillation. It would also have to be determined to what extent natural system damping is available, and take into account other contributions, both positive and negative, to total system damping. Obtaining the necessary data for Southern China may be difficult and it is not known exactly how long this would take. But once these data are in hand, the analysis would not be expected to take more than three to four months.


48

2.5 Economic Feasibility of Increased HEC-CLP Interconnection

Sources of Benefits Under Each Mode of HEC-CLP Interconnection Operation

Firm Capacity — Joint Planning to Defer Construction of New Generation Plant

2.5.1 The major potential sources of benefits from using the interconnection for Firm Capacity would come from reducing the overall cost of investment by increasing investment on interconnection to enable greater mutual sharing of reserve plant, thereby reducing the overall amount of reserve plant capacity required and realising a reduction in investment on generation.

Firm Power — Joint Dispatch to Run the Lowest Cost Mix of Generation Plants

2.5.2 Operational benefits from using the interconnection for Firm Power could arise from: allowing for the dispatch of the most economic combination of

plants on the system as a whole, which would tend to be a lower cost option than the most economic combination of plants in each system separately; and

optimising plant on the shared system to run at more efficient load points.

Approach to Assessment of Economic Feasibility

Economic Benefit to Hong Kong Overall — Benefit/Cost Analysis

2.5.3 The detailed quantitative component of the economic feasibility analysis was conducted by: developing scenarios with generation expansion schedules for

each utility for each interconnection case that would meet the specified (LOLP and operational) planning criteria in each year for each utility;(1)

1(?) The capacity expansion schedules for the interconnection/generation scenarios were developed using Joint Planning principles, treating HEC and CLP from a technical planning point of view as one system comprised of two areas with a capacity-constrained transmission interconnection between them. The key technical difference between the scenarios is the interconnection capacity — the extent of the inter-area transmission constraint — in each year of the analysis. Adopting this standard technical/economic modelling approach allowed the potential for technical and economic benefits of interconnection to be identified. The existing utility structure in Hong Kong is one of two separate companies under separate SCAs. Questions regarding the extent to which the present arrangements may or may not be compatible with integrated approaches to utility planning and operation, and which aspects of those arrangements would need to be changed to allow a more integrated approach, were considered separately in this Study.


49

calculating the investment cash flow for each generation and interconnection investment in each schedule;

calculating the discounted present value of each investment stream for the periods to 2008, to 2018 and to 2028; and

comparing the present values between the scenarios. The economic analysis was conducted in real currency terms and present valued to 1999 using a real discount rate of 4% per annum.

2.5.4 In addition to this, the production cost implications of each interconnection/ generation planning scenario were considered qualitatively, taking into account the relative values of fuel prices for the various existing plants in each utility and the added plants under each planning scenario.

Effect of Economic Benefits on Average Tariffs and Total Customer Bills under the Current Scheme of Control Agreements — Financial Analysis

2.5.5 In any electric utility operation, all revenues come from customer bill payments. Under the SCA accounting framework, economic benefits from lower overall investment schedules (due for example to deferral of investment in fixed assets such as generation plant) translate into consumer benefits for two main reasons: the revenue required to repay the principal and interest on loans

obtained to finance assets is lower when the overall investment cost required is lower; and

the absolute value of the utilities’ permitted return is a proportion of the total value of average net fixed assets, and is therefore lower in absolute terms when the overall investment cost is lower.

Estimated Overall Economic Benefits for Each Interconnection/Generation Scenario

Estimated Overall Economic Benefits from Firm Capacity

2.5.6 The economic results are presented on a consolidated basis, because the first stage of the analysis is technical and economic, it necessarily takes a total resource perspective and therefore does not take into consideration the division of ownership between areas within the interconnected system. Figure 2.5.a shows the consolidated HEC and CLP results for the present value of incremental capital expenditure on generation and interconnection investments (discounted to 1999 at 4% per annum) associated with each of the scenarios. The discounting analysis is based on the expected schedules of capital expenditure in the years leading up to commissioning of each investment under each interconnection/ generation scenario.


50

Figure 2.5.a Discounted Present Values of Cumulative Incremental Capital Expenditures for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP

0

5 000

10 000

15 000

Base Scenario A Scenario B Scenario C

Interconnection / Generation Scenario

Capi

tal E

xpen

ditu

re (

PV 1

999

HK$

mill

ion)

to 2008

to 2018

to 2028

Discounted at 4% per annum real

2.5.7 The figure shows that: each of the three scenarios has somewhat lower relevant

incremental capital expenditures than the base case, but those in Scenario B, whereby the HEC-CLP interconnection is upgraded by the addition of 2*550 MVA 275 kV cables in 2004 would provide the lowest incremental cost capital investment schedule overall;

Scenario B has lower incremental capital expenditures in general (the present value of the costs up to 2008 is very slightly lower in Scenario C, but that scenario has higher costs in subsequent periods); and

the present value of the overall savings on costs with Scenario B is about five per cent of the present value of the relevant costs of the base case.


51

Table 2.5.a Discounted Present Values of Cumulative Incremental Capital Expenditures for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP $ million, present values to 1999 discounted at 4%per annum (real)

Base Scenario A Scenario B Scenario CTotal Net Present Value of Each Investment Stream to 2008 6 725 6 616 6 378 6 289to 2018 12 560 12 292 11 999 12 244to 2028 14 549 14 309 13 653 14 086Absolute NPV Saving between Base and Each Interconnection/Generation Scenarioto 2008 - 109 347 436to 2018 - 268 562 316to 2028 - 240 896 463NPV Saving as a Proportion of Base for Each Interconnection/Generation Scenarioto 2008 - 2% 5% 6%to 2018 - 2% 4% 3%to 2028 - 2% 6% 3%

Note: percentage savings are a proportion of the relevant incremental costs in the base case. Inclusion of other incremental costs in the base case would lower the percentage values.

2.5.8 This result shows that interconnection/generation Scenario B is the most attractive from a purely economic point of view.(1) The values in Table 2.5.a show that the cumulative discounted present values of incremental capital expenditure (on interconnection and generation) in Scenario B would be lower than the comparable value in the base case by $347 million, $562 million and $896 million in the periods up to 2008, 2018 and 2028 respectively. Further analysis is presented in the following sub-section which shows that under the accounting approach embodied in the existing SCAs, if: the financial effects were consolidated and spread evenly

between HEC and CLP customers; and the reductions in the SCA annual revenue requirements

associated with the estimated potential economic savings in each year of the period up to 2008 were spread evenly over the number of units (kWh) of energy projected to be sold in each year of the period up to 2008;

then the average of those annual, consolidated tariff reductions for the period up to 2008 would be about 0.4 ¢/kWh.(2)

1(?) Thus the economic and technical results are consistent, interconnection Scenario B being the most sound of the scenarios from a technical point of view, for the reasons explained in the previous sub-section. 2(?) There may be additional savings available later from subsequent interconnection options not analysed in this study, which would change these results slightly, but the magnitudes of the values involved would be expected to be similar.


52

Estimated Overall Economic Benefits from Firm Power

2.5.9 Consideration of joint economic dispatch could only increase the economic benefits of each of the interconnection/generation scenarios. However, the potential benefits for the period to 2008 of joint economic dispatch under Firm Power use of the interconnection would be small relative to the Firm Capacity benefits quantified above. This is due to the following technical, contractual, operational, economic, and environmental constraints on Joint Dispatch in the period to 2008. Technical considerations mean that use of the interconnector for

Firm Power transfer would require a stable system free from low frequency oscillation problems. As explained in the technical considerations above, Firm Power arrangements are not considered advisable before this problem is resolved. Firm Power could therefore not be implemented before 2004 at the earliest.

Contractual constraints on dispatch throughout the period to 2008 would apply due to CLP’s commitment to take electricity from the Daya Bay nuclear plant, which would be treated as must-run for dispatch purposes.

Contractual constraints on dispatch throughout the period to 2008 would apply due to CLP’s take-or-pay commitment to buy gas for the Black Point combined cycle gas turbine plants.

Operational constraints to potential benefits from Joint Dispatch include Hong Kong’s low load diversity — the load profiles of HEC and CLP are similar and the two areas experience peak demand at similar times. This limits the potential for operational synergy between the two areas.

Economic constraints include the low fuel diversity of marginal plants in Hong Kong, given the (imported nuclear-generated and gas) take-or-pay contracts and the fact that all coal is purchased on world markets by both CLP and HEC at what must therefore be assumed to be the same average price over time.

Environmental constraints would likely require the preferential dispatch of the HEC coal plant fitted with flue gas de-sulphurisation for environmental reasons, otherwise that plant would be at a cost disadvantage in the economic dispatch merit order.

Given these constraints, the magnitude of the economic benefits from Joint Dispatch would likely be quite limited, at least in the short- to medium-term.

2.5.10 In the medium- to long-term (after expiry of Daya Bay contract in 2014, and Black Point take-or-pay contract in 2015) more flexible opportunities for Joint Dispatch may exist, but modelling these now would require very broad and sweeping assumptions about fuel prices and contracts in the distant future.


53

2.5.11 Joint Dispatch would be required under a future competitive wholesale market (under the Multiple Wholesale Purchasers model and to some extent under the Single Wholesale Purchaser model discussed in Section 3).(1) Modelling at this time economic dispatch arrangements for the medium- to long-term future — by which time a completely different market structure may be in place — would not provide a great deal of useful information as it would simply be the product of very broad and sweeping assumptions about the distant future.

2.5.12 Some additional points with respect to Joint Dispatch should also be noted. Joint Dispatch, or use of the HEC-CLP interconnection for Firm

Power, would require a substantial re-negotiation of the voluntary Interconnector Agreement between HEC and CLP, as the current agreement requires that neither utility install generation capacity with the intention of supplying the other parties’ load.

In fact, the production costs would vary between the interconnection/ generation scenarios, because the generation mix would vary between scenarios. The most significant factor here is that the Lamma extension generation units are planned to run on natural gas piped from an LNG receiving terminal in Southern China. HEC have indicated that purchase of this fuel would require entering into a take-or-pay contract. Therefore the Lamma extension units would be designated as “must-run” in the dispatch scheme. Since natural gas piped from the LNG terminal would be more expensive than the other fuels in the generation mix, this would tend to increase the production costs for the system as a whole. Therefore, in scenarios where the Lamma extension is delayed, the present value of production costs would be lower. In this way, the economic benefits of interconnection/generation Scenario B would be increased by dispatch considerations, even before consideration of potential benefits from Joint Dispatch under Firm Power arrangements. However, since the future contract price for such gas is currently not known, these benefits are not precisely quantifiable.

Comments on the Results of the Economic Analysis

2.5.13 The results of the economic analysis are: dependent on the estimates of the costs of the generation plant

and the interconnection upgrades; generally economically conservative and both technically and

economically conservative in one respect; and

1(?) In a competitive generation environment, electricity generators may still need to enter into long-term fuel purchase agreements, however the responsibility for the commercial risk associated with such fuel selection and sourcing decisions would then rest with the management of the investor parties, as is the case in making supply contracts in other types of competitive markets and in other competitive electricity markets around the world.


54

to some extent the result of costs already sunk in existing assets in the interconnected system.

These points are each discussed in turn below.

Effect of Cost Estimates

2.5.14 If the cost of increased interconnection capacity was higher than estimated, the benefits would be lower than indicated. Conversely, if the costs were lower, the benefits would be higher. If generation plant was more expensive than estimated, the benefits would be higher. Conversely, if the generation plant was less expensive than estimated, the benefits would be lower.

Effect of Conservative Assumption

2.5.15 In the calculation of the results presented in Table 2.5.a it is assumed that it is not possible to defer the construction of the sub-marine transmission cables from Lamma Island to Ap Lei Chau, even if the Lamma Extension plant is deferred beyond 2004. The construction of these cables would: allow the output from generation plant at Lamma to be

maximised in supplying HEC load; ensure that the Lamma-Ap Lei Chau transmission capacity would

not trigger the need to implement Firm Power arrangements for use of the HEC-CLP interconnector (which would require renegotiation of the Scheme of Control and Interconnector Agreements) before such a need was triggered by HEC load exceeding generation capacity, which in Scenario B would occur after the existing SCAs expire in 2008;

ensure that there is no possibility that the results of the LOLP analysis would be adversely affected by Lamma-Ap Lei Chau transmission capacity constraining the supply path from Lamma generation to CLP load, rather than HEC-CLP interconnection capacity as allowed for in the LOLP analysis.

2.5.16 Further detailed analysis would be required to confirm the exact timing necessary for the installation of the additional Lamma-Ap Lei Chau transmission cables. Such an analysis could form part of a comprehensive Level I and Level II reliability review and analysis for the interconnected system as a whole, as discussed in paragraph 2.3.18. If such an analysis showed that: deferral is possible from the perspective of integrated Level I and

Level II reliability for the interconnected system; and

the Scheme of Control and Interconnector Agreements could be renegotiated to allow use of the HEC-CLP interconnection for Firm Power transfer; and

the LOLP values calculated for future years in the analysis underlying Scenario B would not be adversely affected by the


55

constraint of the existing Lamma-Ap Lei Chau transmission capacity;

then the long-run potential economic benefits of deferral would increase by some $350 million. The Consultants’ experience suggests that these potential economic benefits could be captured, but have not included them in the main analysis on the basis that a conservative under-estimate of economic benefits is preferable to an over-estimate.

Effect of Costs Already Sunk in Existing Assets

2.5.17 The results of the economic analysis presented above are to a certain extent due to the existence of investment already made in site formation and the transmission network. In the hypothetical case where these costs were not sunk, investments of up to $400 million(1) present value would be incurred in Scenario B relative to the base case in the period to 2018.(2)

2.5.18 The analysis conducted implicitly assumes that there is sufficient additional gas in or near the present field in the South China Sea and sufficient pipe capacity to supply generation plant at CLP beyond C8. This is a qualitatively different issue from the effect of sunk costs, because it is about potential additional incremental costs in future, but the sensitivity of the result to this assumption can be examined in a similar way to the sunk costs. If this assumption was dropped and it was assumed instead that the only option for gas to supply post-C8 CLP generation units was from a future LNG terminal in Shenzhen (same source assumed for the future HEC Lamma Extension), then the reduction in the present value of economic benefits from Scenario B relative to the base case would be about $15 million, arising from a difference in timing of investment in the pipe from the future LNG terminal.

Environmental Considerations with Respect to Scenario B

2.5.19 Under scenarios where the natural gas fired Lamma plant is delayed, there may be an increase in emissions from electricity generation if the output from coal plant from CLP is thereby increased relative to the base case where Lamma generation plant is not delayed.

1(?) This value is based on assumed costs and assumptions about when the assets would be needed if they were not already in place. 2(?) The discussion of this topic focuses on values for the period up to 2018, rather than up to 2008 to avoid present values representing deferral of investments beyond the 2008 dividing line and into the decade to 2018 appearing as if complete avoidance of the investment had occurred.


56

Effect of the Economic Results on Each Individual Company and Its Customers

2.5.20 It is noted above that the economic results presented are consolidated between HEC and CLP, as the system was considered from an integrated perspective in the technical and economic analysis. In practice, the inter-utility operational implications of the generation/interconnection scenarios would need to be considered to ensure that — under the existing SCAs — each utility is satisfied that the arrangements in place allow it to fulfil its obligations to ensure that its customers have reliable electricity supply. Under the interconnection/generation scenarios in this Study, one company would rely to a significant extent on generation capacity support from the other company. HEC have indicated that they would be very concerned about meeting their customer loads reliably under such an arrangement, whereby most of the reserve generation capacity would be located outside their system. At the very least, substantial changes to the Interconnection Agreement (and possibly the SCAs also) would be required for such an arrangement to work to the satisfaction of both companies.

2.5.21 The fact that the Hong Kong electricity supply system is comprised of two separate companies that are separately regulated also affects the way that the potential economic benefits identified would flow through to the revenue requirements for each utility, and how they would affect tariffs and customer bills in each individual utility. These considerations are analysed in the following sub-section.

2.6 Consumer Bill and Tariff Impacts of Increased HEC-CLP Interconnection

Sources of Consumer Benefits under the Existing Scheme of Control Agreements

2.6.1 The economic savings calculated from the investment streams under each of the interconnection scenarios described above would on average:(1) reduce the total amount of revenue required by the utilities to

meet their costs; reduce the value of average net fixed assets; and therefore reduce the absolute amount of return permitted to the utilities

under the Scheme of Control rate-of-return formula.(2) 1(?) The words “on average” refer to both the overall picture — on a consolidated HEC/CLP basis — and the effect over time. The details — when each utility is considered separately and when the time effects are considered are somewhat more complex and are discussed separately. 2(?) Note that the results of the financial and tariff analysis presented in this sub-section show the effect of the scenarios analysed on revenue requirements and tariffs that would occur if the existing SCAs were to remain in place throughout the entire time frame to 2028. Any future modifications to the accounting framework embodied in the existing SCAs would change the way that the economic results translate into financial and tariff results from that presented here.


57

The economic savings would therefore translate on average into customer bill savings and tariff reductions.

Approach to Assessment of Consumer Benefits

2.6.2 The overall potential economic savings described above would on average translate into customer bill savings and tariff reductions. The present value of customer bill savings depends to some extent on: the ratio of debt-to-equity used to finance new assets; the interest rate at which the utilities borrow capital to finance

assets; the repayment schedule for the loans; the depreciation schedule; the inflation rate used for prices of capital equipment; and the nominal discount rate used to express the savings in present

value terms.

2.6.3 The financial analysis was conducted in nominal currency terms and present valued to 1999 using a nominal discount rate of 8% per annum. Additional assumptions adopted were that: CLP finances new assets with 20% equity and 80% debt with the

principal repaid in equal instalments over ten years; HEC finances new assets with 100% debt, with the principal

repaid in equal instalments over thirty years; assets are depreciated on a straight line basis over thirty years;

and the interest rate on borrowed capital is 8% per annum. A real discount rate of 4% per annum was adopted, to be consistent with the economic analysis.

2.6.4 For financial analysis using market interest rates, it is necessary to take account of inflation in future asset values to obtain a correct result. The combination of a nominal discount rate of 8% per annum and a real discount rate of 4% per annum imply an annual inflation rate for prices of fixed assets of 3.85% per annum.(1) Although inflation is low in the current economic downturn, Hong Kong has experienced sustained periods of quite high consumer price inflation. In that context, the implied long-term inflation value used here may appear low. However, consideration of the following points suggest that the value is not too low:

1(?) The inflation rate is implied from the real and nominal discount rates and calculated as follows. Let n, r and i be the nominal, real and inflation rates respectively. Therefore (1+n) = (1+r).(1+i). Hence the inflation rate, i = (1+n)/(1+r) -1.


58

the price inflation of the assets under consideration are determined largely by world market rates for the equipment, not by local Hong Kong inflation;

the cost of combined cycle gas turbines is the most significant component in the investment schedules;

the prices for combined cycle gas turbines have fallen quite substantially in recent years as the market for supply of these units has matured and as a result of competition in the international market for sales of these units, so it would be unreasonable to use a high figure for price inflation of these units over a long time frame; and

there is no reason to expect that prices of transmission system equipment on the world market would escalate rapidly.

Estimated Effect on Revenue Requirements, Customer Bills and Tariffs Under the Current SCAs

2.6.5 The calculated effects on utility SCA annual revenue requirements of Joint Planning under each scenario are graphed in Figure 2.6.a and shown Table 2.6.a. These effects would flow through to customer bills via tariffs. A reduction in revenue requirement appears as a negative number in the figure and would be associated with customer bill savings. The savings — plotted in the graph as annual reductions in revenue requirement, and included in the lower portion of the table in net present value terms for each of the three periods — represent the difference between the incremental investments of each scenario relative to the base case scenario. For example, the present value up to 2008 of the Scenario B line in Figure 2.6.a appears in Table 2.6.a as the number PV$1 236 million. It is important to note that these results represent the incremental effect of new generation and interconnection investments in each scenario.

2.6.6 These results show that: the reductions in annual revenue requirement are most significant

in the early years; there would be some increases in utility annual revenue

requirements in the periods to 2018 and to 2028; these would slightly off-set the reductions in the early years, but

the overall revenue requirement would be reduced such that customer bill savings would remain positive, an effect reinforced in the table because the later (negative) values are more heavily discounted than the early (positive) values.


59

Figure 2.6.a Annual Differences in SCA Revenue Requirements Between Each Interconnection and Generation Planning Scenario and the Base Case Consolidated Across HEC and CLP

-400

-300

-200

-100

0

100

200

300

400

1998

2008

2018

2028

Cons

olid

ated

HEC

& C

LP S

CA R

even

ue R

equi

rem

ent

Rela

tive

to

Base

Cas

e (1

999

$ M

illio

n)

Scenario AScenario BScenario C

2.6.7 When interpreting these results, it is important to bear in mind a number of factors: the results in the early years are more certain than the results for

the later years; there may exist opportunities for further increased

interconnection capacity in the period to 2018 and 2028, which would again deliver economic benefits and consumer savings, but these have not been assessed in detail in this feasibility Study.


60

Table 2.6.a Discounted Present Values of Cumulative SCA Revenue Requirements for Interconnection and Generation Planning Scenarios Consolidated Across HEC and CLP $ million, present values to 1999 discounted at 8%per annum (nominal)

Base Scenario A Scenario B Scenario CIncremental Net Present Value of Each Investment Stream to 2008 4 184 3 779 2 948 2 897to 2018 13 356 13 005 12 361 12 704to 2028 20 118 19 802 19 194 19 833Absolute NPV Saving between Base and Each Interconnection/Generation Scenarioto 2008 - 405 1 236 1 287to 2018 - 350 994 652to 2028 - 316 924 285

2.6.8 These reductions in revenue requirements would not be evenly distributed between HEC and CLP. For example, if Scenario B was implemented under the existing Scheme of Control Agreements without any change or adjustment to those agreements, then in the period to 2008 HEC would experience a decrease in annual revenue requirements relative to the base case with a present value of $2 003 million, while CLP would experience an increase in annual revenue requirements relative to the base case with a present value of $767 million.

2.6.9 The effect on revenue requirements translates to customer bill savings via tariff changes. This effect on tariffs, if spread evenly across all projected units (kWh) of energy sales and consolidated between HEC and CLP is shown in Figure 2.6.b. The graph plots the annual tariff in each scenario relative to the base case. Negative values represent a tariff reduction, positive values represent a tariff increase. The graph follows the same general pattern as the annual revenue requirements, due to the form of revenue and tariff accounting embodied in the SCAs.

2.6.10 The impacts are small relative to the total tariff. In practice, transfers to and from the Development Fund provided for under the Scheme of Control could be used to smooth out tariff adjustments over time and to adjust for the combined effect of the time value of tariff revenues and changes in total energy consumption from year to year to meet the revenue requirement in each year. Taking a simple average for illustrative purposes of the annual tariff effects for Scenario B over time, the tariff decrease values would be 0.40 ¢/kWh, 0.14 ¢/kWh or 0.09 ¢/kWh if taken over the periods up to 2008, up to 2018 or up to 2028 respectively.(1)

1(?) If the three periods were taken separately, then the average tariff relative to the base case would be -0.40, +0.08 and +0.02 ¢/kWh in the three successive periods. When combined with the effect of the projected energy sales in each year, these would amount to an average tariff reduction over the entire period to 2028 of 0.09 ¢/kWh.


61

Figure 2.6.b Annual Tariff Differences Between Each Interconnection and Generation Planning Scenario and the Base Case Consolidated Across HEC and CLP

-1.0

-0.5

0.0

0.5

1.0

1998

2008

2018

2028

Annu

al Im

pact

s on

Tar

iffs

Rela

tive

to

Base

Cas

e (1

999

c/kW

h)

Scenario AScenario BScenario C

2.6.11 However, since the change in revenue requirements under the SCAs for each scenario relative to the base case would be unevenly distributed between the two utilities, particularly in Scenario B, the tariff effects would also be different between the two utilities. Figure shows the annual tariff effects for Scenario B for each utility separately as well as the consolidated effect that is also shown in Figure 2.6.b.

Figure 2.6.c Annual Tariff Differences Between Interconnection/Generation Scenario B and the Base Case Consolidated Across HEC and CLP and for HEC and CLP Separately

-4

-2

0

2

1998

2008

2018

2028

Annu

al Im

pact

s on

Tar

iffs

Rela

tive

to

Base

Cas

e (1

999

c/kW

h)

HECConsolidatedCLP


62

2.6.12 Again using the simple average approach for illustrative purposes, this time to compare the effect on each utility’s customers separately under the SCA, the overall savings and overall reductions in revenue requirements would translate under the existing SCAs into: average tariff reductions for HEC customers of 2.21, 1.24 or

0.72 ¢/kWh if spread over the periods up to 2008, up to 2018 or up to 2028 respectively; and

average tariff increases for CLP customers of 0.35, 0.31 and 0.17 ¢/kWh if spread over the periods up to 2008, up to 2018 or up to 2028 respectively.

Estimated Effect on Average Tariffs and Total Customer Bills from Firm Power

2.6.13 Consistent with the explanation in the discussion of potential benefits available from Firm Power above, the customer savings would increase when production costs are taken into account and when Joint Dispatch of plant is considered, but realisation of such benefits is not considered possible before 2004 at the earliest.

Comments on the Results of the Financial Analysis

2.6.14 The results of the financial analysis show that under the arrangements of the present Scheme of Control, the overall potential economic benefits of increased interconnection would result in a reduction in the revenue requirements under the existing SCA accounting framework, which would flow through via tariff reductions into bill savings for Hong Kong consumers on a consolidated basis.

2.6.15 However, the distribution of these economic benefits, both between HEC and CLP and between HEC’s customers and CLP’s customers would not be even. Under Scenario B — the most favourable from an overall economic perspective — HEC customers would experience lower bills and tariffs than they would under the base case scenario (equivalent to lower revenues for HEC) and CLP customers would experience slightly higher bills and tariffs than they would under the base case scenario (equivalent to higher revenues for CLP). This arises from the fact that the Scheme of Control Agreements determine utility revenues as a proportion of fixed assets and under Scenario B, CLP would have more fixed assets and HEC considerably less than in the base case, with the total value of fixed assets being less than the base case.

2.6.16 This imbalance in the distribution of economic benefits that would emerge from the implementation of increased use of interconnection without any simultaneous change in or adjustment to the present Scheme of Control Agreements is a reflection of the point noted in Section 1 Introduction that the existing regulatory arrangements


63

were conceived when the two Hong Kong utilities were not interconnected at all. The establishment in 1981 of interconnection between the two utility systems can be seen as the first step in joint planning. However, the existing institutional arrangements do not fit naturally with complete development of least-cost capital expenditure options that would require extensive co-operation between two separate and separately-regulated companies.

2.6.17 Both utilities have expressed concerns that relate to these wider implications of the interconnection/generation scenarios in the context of the present Scheme of Control and Interconnection Agreements. Such concerns relate to the commercial interests of each electricity company and what they see as their contractual rights under their existing Scheme of Control Agreements with Government.

2.6.18 CLP have a legally binding contractual commitment with a supplier to purchase the final two units for the Black Point power station, based on previous Government approvals. CLP have stated that they have already proposed to Government to exercise to the maximum extent possible the deferral option in their contract with the supplier for these units and that there is no further deferral option allowed in the existing contract. Such commercially confidential data on penalty clauses were not available to the Consultants and the timing of installation of these units in each interconnection/generation scenario was modelled in this feasibility Study strictly on a technical “as needed” basis, rather than on a contractual commitment basis and no deferral penalty costs were included in the analysis. Modelling the two units at zero incremental cost because they are contractually committed and including penalty costs for their deferral beyond the contractual deadline would increase the overall economic benefits of Scenario B relative to the base case. This effect on the overall economics would apply regardless of whether the penalty costs were borne by the utility, by consumers or some combination of the two under the Scheme of Control Agreement.

2.6.19 However, as the financial results presented above show, the installation of a larger interconnector and making increased use of HEC-CLP interconnection to enable system-wide economic benefits to be captured by the installation of Black Point units 7 and 8, and subsequent units as in Scenario B, would result in slightly higher asset levels for CLP relative to the base case in each given year, and significantly lower asset levels relative to the base case for HEC in all years. This would therefore significantly disadvantage HEC from a commercial point of view under the structure embodied in the existing Scheme of Control Agreements, whereby utility profits are proportionately related to the value of each utility’s fixed assets.

2.6.20 The Consultants can appreciate the merits of the points of view of both companies, as well as the opportunities available for economic and consumer benefits. The challenge arising from this Study for


64

the two utilities and for Government is to find the most appropriate resolution of these conflicting requirements.

2.7 Key Conclusions on Interconnection

2.7.1 It is considered technically feasible to increase interconnection capacity between HEC and CLP. Benefits in Hong Kong are limited by the similarities of peak time and generation plant types between the two systems, but overall economic benefits from increased interconnection could be realised by jointly planning generation expansion to share reserve plant capacity. Additional potential benefits from optimising plant operation between the two systems are considered to be small.

2.7.2 Under the existing Scheme of Control Agreements, the overall economic benefits from increased interconnection would be distributed unevenly between the two electricity companies and their customers, since the investment in generation and transmission assets would be distributed asymmetrically.

2.7.3 In general, it is observed that there is a clear disjunction between realising potential economic benefits from the combined utility systems and the financial incentives arising from both the form of regulation under the existing Scheme of Control framework, and its separate application to individual utilities. In the near-term, a scenario that seeks to realise the maximum overall potential economic benefits would further exacerbate this effect, because it would require more fixed assets to be added to one company than the other.

2.7.4 To be certain of meeting the projected 2004 HEC summer peak load with the required level of reliability, the adjustments or modifications of the Scheme of Control Agreements necessary for the additional interconnection option would need to be completed by the end of 1999 or early 2000. While this may be theoretically possible, it seems very unlikely that it could be accomplished in practice. The ability of the system to meet the 2004 HEC peak load with the desired level of reliability would be contingent on resolution of these complexities within a very short time-frame. The critical decision-point is set by the longer of the two lead times to provide either increased interconnection or additional generation capacity within HEC to meet the 2004 peak load. When this point is reached, a decision will need to be made one way or the other to ensure reliability of supply is not compromised.


65

3 FEASIBILITY OF THE INTRODUCTION OF ELECTRICITY SECTOR COMPETITION IN HONG KONG

3.1 Introduction

3.1.1 Around the world many governments and regulators are restructuring and introducing competition to their electricity sector, or are in the process of doing so. This process has been undertaken to a greater or lesser extent in countries including the United Kingdom, Norway, New Zealand and Australia. Some states of the United States of America, such as California, have already introduced electricity sector competition. The process of introducing competition to the electricity industry at a Federal level is still in its early stages. In a number of places with a previously state-owned electricity sector, including England, Australia and New Zealand, this process has also involved sale to private interests of at least some parts of the electricity industry.

3.1.2 In most cases the aim of these reforms has been to increase efficiency, based on the view that private companies, acting more or less freely within the discipline of competitive markets, make better investment decisions, make better use of existing plant, better manage their business and provide better choices for customers, than do either state-owned monopoly enterprises or privately-owned companies with virtual monopolies operating under Government regulation. Experience of electricity competition is still relatively short, particularly as it has been phased in over a number of years in most places, and adjusted to overcome initial difficulties in other places. Conclusions to date on the merits of electricity competition in general have tended to depend to a large extent on acceptance of the view of competitive markets just described.

3.1.3 In some cases pressure for reform has come from customers who feel they can purchase more cheaply elsewhere; in other cases the incumbent utility has become inefficient and the rationale is to introduce incentives; in yet other cases the utilities themselves want to be freed from what they see as inhibiting regulatory intervention.

3.1.4 The Government of the HKSAR is currently seeking to establish whether encouragement of competition in the Hong Kong electricity supply sector would be in the interests of consumers. The remainder of this report sets out the opportunities that are available for the development of the market and proposes and evaluates likely options.

3.1.5 The next section (Section 4) discusses likely market developments in Mainland China.


67

3.1.6 It should be noted that the terms electricity sector, electricity industry, electricity supply industry, utility industry and utilities are often used interchangeably in the industry’s literature. For the purposes of this report they are considered to include the activities that involve the generation, transmission, distribution and retailing of electricity.

This Section

3.1.7 This Section is divided into sub-sections which: summarise the local and wider context for future introduction

of electricity sector competition in Hong Kong, discuss the current Hong Kong situation and reasons why introducing competition to Hong Kong might be considered, outline various market structures and briefly review the experience of other countries in introducing electricity competition, provide an overview of electricity competition in general and identify which specific aspects of particular international experience of electricity sector reform and the introduction of competition may allow for relevant comparisons with Hong Kong;

describes four market structures and how each might be applied to Hong Kong, including alternative forms of regulation for the non-competitive parts of the industry, to determine the most appropriate options for Hong Kong;

evaluates in detail the most appropriate options for Hong Kong against a specified set of criteria, examines detailed variations of each market structure, assesses the most appropriate variation for Hong Kong within each broad structure, describes how each of these might be applied in Hong Kong, indicates a set of market transition options starting from the existing structure and indicates possible time-frames for implementation in terms of earliest possible dates for each type of market structure, for assessment by policy makers.

3.2 Local and Wider Context for Electricity Sector Competition

3.2.1 This sub-section considers: the development of the Hong Kong electricity sector; the current Hong Kong situation in the electricity sector; current institutional arrangements in the Hong Kong electricity

sector; policy background and motivations for the consideration of

competition in the electricity sector; general advantages and disadvantages of electricity sector

competition; alternative electricity market structures; and ERM GOVERNMENT OF THE HKSAR

68

comparisons with electricity sector reform and the introduction of competition internationally and their applicability to Hong Kong.

Development of the Hong Kong Electricity Sector

3.2.2 The two existing privately owned utilities — operating in more recent decades under the Scheme of Control Agreements — have provided Hong Kong with an electricity sector that has reliably met the demands of the territory throughout periods of dramatic economic growth. Comparisons indicate that today’s prices for electrical power are somewhere in the middle of the range internationally, based on simple, aggregate comparisons. Initial observations of the Hong Kong utility operations do not suggest that there are significant inefficiencies within the utility operations, as was the case with a number of the state-owned utilities in other countries prior to privatisation and the introduction of competition.

3.2.3 As indicated previously, the current two-utility structure is the result of the historical development of the electricity industry in Hong Kong. Indeed the Scheme of Control Agreements originated in the 1960s and 1970s. Table 3.2.a shows that there are a number of characteristics of Hong Kong, Mainland China and the electricity sector that have changed in recent times, and that will continue to develop. These changes suggest that a review of the industry structure and the form of regulation allowing time to plan changes carefully before the current SCAs expire in 2008 may be timely.

3.2.4 It is important that the electricity sector continues to meet Hong Kong’s needs as those needs change and grow. Consideration therefore needs to be given to the current situation and expected developments. Hong Kong, Mainland China, their electricity sectors and the electricity supply industry internationally have changed and developed significantly in recent times, in the following ways:

3.2.5 Continuing change is expected in a number of areas, including: continued rapid development in the size of the Southern China

system; some factor costs of electricity production are expected to be

lower in Mainland China than Hong Kong (particularly land for sites and labour);

although some of these potential advantages may be off-set by the complications of the PRC taxation system and concerns about differences in environmental controls on electricity generation between Hong Kong and Mainland China.


69

Table 3.2.a Background Situation in Hong Kong and Mainland China

Situation in Hong Kong and Mainland China until recent times

Current Situation and Expected Future Developments

Hong Kong was a separate political entity from Mainland China, under separate administration

Hong Kong is now part of Mainland China under the “one country, two systems” arrangement

Hong Kong was entering a period of unprecedented economic growth and growth in demand for electricity

growth in demand for electricity in Hong Kong is not as dramatic as it was in the early previous decades of industrialisation, and Hong Kong has now de-industrialised and entered a qualitatively different phase of economic growth

a stable environment to encourage the necessary investment in electricity infrastructure was needed

the two Hong Kong utilities are now more mature, modern companies, with a strong history of investment

HK had two separate, isolated electricity systems

the two Hong Kong utilities are now interconnected

CLP is quite strongly interconnected with the Southern China grid

technically and economically feasible opportunities exist to increase the interconnection between HEC and CLP and to make increased use of the interconnection

the electricity system in Southern China was not at all well developed

the Southern China system has grown to around three times the size of the Hong Kong system

electricity systems world-wide were generally either state-owned monopolies or privately-owned monopolies regulated by government

electricity systems world-wide are being re-structured from monopolies to competitive structures

APEC is promoting the investigation of competitive market options in the electricity sector

there were many examples of neighbouring electricity systems which were not interconnected

Governments are increasingly looking to interconnect or increase interconnection between neighbouring electricity systems, as well as introducing competition (or as a means to facilitating competition)

the Association of South-East Asian Nations (ASEAN) is currently examining the feasibility of interconnecting its electricity systems

Current Situation of the Hong Kong Electricity Sector

3.2.6 CLP(1) and HEC are the only companies which have invested in electricity generation, transmission and distribution in Hong Kong. The companies are investor-owned and are effectively the sole

1(?) Esso Eastern Inc., part of the Exxon corporation has made joint, co-operative investments with CLP in electricity generation. For the purposes of the Scheme of Control Agreement and for this Study, they may be considered to operate as an integrated whole.


70

suppliers of electricity in the areas that they serve (refer to Market Structure 1, Vertically Integrated below).

3.2.7 CLP provides power to the Kowloon peninsula, Tsing Yi Island, Lantau Island (including the new airport), the New Territories and a number of small islands. HEC serves Hong Kong Island and the neighbouring Islands of Ap Lei Chau and Lamma. Additionally, several small village co-operatives produce electricity for some remote localities.

Current Institutional Arrangements in Hong Kong

Scheme of Control Agreements (SCAs)

3.2.8 The Hong Kong Government regulates the electricity-related activities of the two companies through separate Scheme of Control Agreements (SCAs), which constitute the major element of the regulatory framework for the power sector in Hong Kong. They include rate-of-return regulation for the two utilities as well as price control and approval mechanisms. These agreements establish what costs the companies may pass on to consumers, the profits they may earn and effectively give the Government authority to approve the companies’ expansion plans and investment proposals. The current fifteen year agreements will expire in 2008.

3.2.9 The Government entered into the SCAs on mutually agreed terms with HEC and CLP. As stated by Government,(1) the purpose of the SCAs is to: ensure that consumers get a reliable and efficient supply of

electricity at a reasonable price; ensure that the companies’ shareholders get a reasonable return

on their investment, so as to encourage them to continue to make the investments required to ensure sufficient supply of electricity to meet increasing demand;

facilitate the companies’ efforts to raise funds in the financial markets when needed;

ensure that the companies avoid serious financial difficulties; and achieve these objectives with no direct subsidy from public funds

and a minimum of interference.

3.2.10 The SCAs set out the obligations and rights of the companies and of the Government on behalf of consumers. The agreements provide a framework for monitoring the companies’ affairs to protect the interests of consumers. They require the companies to seek the approval of the Executive Council for certain aspects of their financing plans, including projected tariff levels, and the agreement of Government to each annual tariff adjustment.

1(?) Monitoring of Power Companies, July 1998.


71

3.2.11 The SCAs provide for two reviews to be conducted during the tenure of the agreements, one between 1997/98 and the other in 2003. During each review, the Government and the power companies have the right to request modification of any part of the SCAs. The current agreements with CLP/Exxon and HEC/Holdings commenced on 1 October 1993 and 1 January 1994, respectively. Both are scheduled to last for 15 years.

3.2.12 The SCAs provide an agreed basis for determining the profit that the shareholders of the companies can earn each year from generating and supplying electricity. Profit is measured by the net return allowed to shareholders. This ‘net return’ is calculated by first determining the ‘permitted return’, which is 13.5% of the company’s Average Net (ie, depreciated) Fixed Assets (ANFA) plus an additional 1.5% for those assets financed from shareholders’ investments since 1 October 1978 for CLP and since 1 January 1979 for HEC. Deductions are made from the “permitted return” to give the “net return.” These deductions include interest on long term borrowings of up to 8% per annum; a charge at 8% per annum on the average balance of the Development Fund; interest up to a rate of 8% per annum on the funds from growth in consumers’ deposits after 30 September 1998 for CLP/Exxon and after 31 December 1998 for HEC/Holdings; and excess capacity adjustment (if applicable).

3.2.13 It is important that the power companies should have sufficient incentive to invest in the capital assets needed to ensure that supply will meet projected demand, as the lack of a secure power supply is considered unacceptable for a modern city such as Hong Kong. At the same time, however, it is important to prevent over-investment in generation, transmission and distribution facilities as this will lead to tariffs being higher than necessary. Accordingly, the government retains consultants to examine critically the system development plans of the companies whenever they submit proposals for the financing of major new facilities. The SCAs have been amended, effective from 1 October 1998 for CLP and 1 January 1999 for HEC, to provide a mechanism to discourage investment in excess capacity by not counting a portion of the cost of additional units that are deemed to be in excess under the application of a specified test referred to as the Excess Capacity Threshold (ECT) test.

3.2.14 It should be noted that the SCAs are not actually franchises, nor have they ever provided either HEC or CLP with exclusive supply areas. Any company, provided they meet all other legal requirements, would be allowed to generate, transmit and distribute electricity in Hong Kong. There is nothing in the SCAs to preclude others from entering the market to compete with HEC or CLP.

3.2.15 However, given the generation assets and the established transmission and distribution infrastructure which, between HEC and CLP, reaches the entire established customer base, along with the absence of any requirement for the two utilities to grant third parties access to their transmission and distribution networks, it is unlikely


72

that any company would enter the market to compete with them under the current industry structure. The two utilities do not presently have any access to each other’s transmission and distribution networks(1) and currently have a contractual agreement not to supply customers in each other’s area.

Interconnection Agreement

3.2.16 As described in some technical detail in Section 2, the HEC and CLP transmission networks are interconnected via submarine cables in three circuits of 240 MVA each. The submarine cables are owned by HEC and CLP in equal shares and all associated land equipment and cables owned by individual companies.

3.2.17 The Interconnection Agreement (IA) between CLP and HEC provides the necessary details for the technical configuration of the interconnection system, for the operation and maintenance of the system, for metering for service conditions, and for billing and payment. The IA has been reached between the two utilities and has no direct involvement with Government. The original 1981 IA was extended in 1991 by a supplemental agreement between the two companies and has no expiry date, although it may be terminated by not less than one year’s prior notice in writing by one party or the other.

3.2.18 The IA includes a number of stipulations which are of importance for a potential increase of the interconnection capacity between the two systems and an enhanced exchange of power. These include: Ownership. A clear distinction is made between jointly owned

facilities and those facilities which are owned individually. The submarine cables, connected stop joints and associated equipment and civil works are owned jointly; split equally between the two companies. All other facilities are owned individually by the company in whose service area they are located.

Types of power exchange. The Interconnection Agreement comprises three types of power exchange services: Emergency exchange: Supply of power from one company to

the other under emergency conditions, ie when a company is temporarily unable to supply energy to all its customers.

Economy exchange: The option of supplying short-term power in those cases where the supplier can deliver surplus power at lower variable costs than the variable costs at which the buyer can generate electricity.

Firm exchange: All other types of exchange of power not included in the first two categories, basically intended as

1(?) There is a small exception to this — the electricity supply to the MTR system is designed with full redundancy so that if either utility suffered a blackout, the other could continue to supply the whole MTR system. In this sense both companies may technically be considered to have access to each other’s area through the MTR system.


73

exchange of power for a period not less than one day. This is a different definition from the one normally used in the power industry outside Hong Kong, ie power transfer over a period of time on a committed basis that is usually included in the calculation of generation capacity.

Installed generating capacity. The agreement clearly stipulates that neither party shall install excess power generating capacity with the aim of supplying power to customers in the other company’s supply area. To avoid this, each party is responsible for installing sufficient generating capacity to meet its customers’ requirements. As far as extension of installed generating capacity is concerned, co-ordination and discussion between the two companies is required through the Planning Committee.

3.2.19 The IA provides a clear set of arrangements for technical and operational relationships between CLP and HEC for the exchange of power between the two companies as well as on metering, billing and payment. It focuses on the exchange of power in emergency situations. Other types of exchange of energy are covered by the agreement as well, but in effect they play a minor role only.

Why Think About Competition Now?

The Government’s General Policy on Competition

3.2.20 The Government has set out its wider objectives on competition policy, in a statement issued last year (1) . The statement says:

3.2.21 The objective of the Government’s competition policy is to enhance economic efficiency and free flow of trade, thereby also benefiting consumer welfare. The Government is committed to competition as a means to achieving the said objective, and not as an end in itself.

3.2.22 The Government considers competition is best nurtured and sustained by allowing the free play of market forces and keeping intervention to the minimum. We will not interfere with market forces simply on the basis of the number of operators, scale of operations, or normal commercial constraints faced by new entrants. We will take action only when market imperfections or distortions limit market accessibility or market contestability, and impair economic efficiency or free trade, to the detriment of the overall interest of Hong Kong. We will strike the right balance between competition policy considerations on the one hand, and other policy considerations such as prudential supervision, service reliability, social service commitments, safety, etc, on the other.

1(?) Statement on Competition Policy, Competition Policy Advisory Group, Trade and Industry Bureau, May 1998.


74

Electricity Prices

3.2.23 Electricity prices vary substantially around the world because of different tariff structures, cross subsidisation and different cost structures. A simple comparison of average net tariffs (total revenues excluding any taxes divided by total energy sales) with other countries suggests that Hong Kong’s utilities provide electricity at rates in the middle of the range internationally. Case Study 1 contains more information on projecting prices in competitive markets.

General Advantages and Disadvantages of Electricity Competition

3.2.24 Electricity sector competition has the potential to: encourage utilities to improve service and offer new

innovative services to consumers to win their business; encourage more cost-reflective pricing structures; relieve Government of the burden of detailed direct

regulatory review of all aspects of utility operations, demand forecasting and of the need to grant approval to each major investment decision required to protect consumer interests as is the case with private monopolies;

automatically determine margin levels through the kind of competitive processes that operate in markets for other goods and services, rather than requiring returns to be set externally by a government regulator; and

reduce prices to consumers as a result of competitive pressures; encourage evaluation of new plant procurement and investment

decisions to reduce the overall cost of the supply of electricity rather than just the construction cost,

drive down costs of building new plants (capital expenditure), maintaining and operating the plants to generate electricity (O&M and, in the long run, fuel costs), and operating expenditures in retailing energy (the costs of metering, billing and serving customers);

drive down margins in generation by competition for contracts (medium-to-long-term) and competition for dispatch (short-term) and in energy retailing by competition for customers;

reward the more efficient operators, who have lower costs and can therefore obtain higher margins at the same price as their competitors;

encourage innovation and new technology, particularly where it offers ways to reduce costs; and

encourage investment by new operators if opportunities exist to operate more efficiently than existing companies.


75

3.2.25 However: obtaining the benefits that results from competitive pressures

requires that the market operates in a way that actually creates these competitive pressures;

competition is not possible in all parts of the electricity sector, in particular transmission and distribution networks are generally considered natural monopoly functions and most jurisdictions with competitive electricity markets continue to regulate these activities;

particularly in small markets, it may be difficult to maintain economies of scale and a sufficient number of companies to provide the competitive market pressures required to deliver the benefits described above;

competition would tend to increase transaction costs, which may slightly reduce the benefits described above;

in competitive markets consumers still consider that the Government has some responsibility to ensure that their interests are protected, and to ensure that a competitive market operates effectively to deliver the benefits described above may involve more complexity than traditional regulation, requiring careful analysis of competitive market behaviour and judgement about how best to maintain or restore competitive pressures;

as competitive pressures increase, so does uncertainty and commercial risk, so companies in competitive markets would tend to have higher costs of capital (higher interest rates) than utilities operating as monopolies with regulated rates of return.

Competitive Electricity Market Structures

3.2.26 Four different market structures are described in this Study, each illustrating different degrees of competition. These four market structures may be thought of as points on a spectrum of increasing competition beginning with generation (supply) and moving down to retail consumption (demand). After the monopoly status quo, this scheme shows increasing levels of competition from active competition between wholesale sellers, then adds active competition between wholesale buyers, then adds the active market participation of retail buyers.

3.2.27 It is possible to adopt alternatives to these structures, or to proceed incrementally in the opposite direction, from the retail/consumption end up to the wholesale/generation end. Indeed New Zealand followed this bottom-up approach to the reform of its electricity sector, largely due to the historical fact of having many local-council based distribution companies and one Government-owned generation and transmission company. This form of competition had limited initial success, as all distributors were essentially buying at the same price from the same supplier. The New Zealand ERM GOVERNMENT OF THE HKSAR

76

Government separated transmission from generation and has more recently split up the generation entity to create a competitive wholesale electricity market. Retail competition without wholesale competition does not really work, and in any case would not be readily applicable to Hong Kong’s existing electricity industry situation.

3.2.28 Each of the four structures considered in detail are described below. Market Structure 1: With monopoly at all levels, generation is

not subject to competition and no one has any choice of supplier — a single company handles the production of electricity and its delivery over the transmission network to distribution companies and/or final consumers. This is often called a Vertically Integrated Market. Separate examples of this market structure can exist side-by-side in separate geographic areas, as is currently the case in Hong Kong. Interconnection between adjoining systems, inter-system exchange or purchase of energy and co-operative arrangements such as sharing of capacity for reserve, emergency support and other purposes can also occur within this structure, as is also currently the case in Hong Kong to a certain extent.

Market Structure 2: A Single Wholesale Buyer or purchasing agency that chooses from a number of different generators, to encourage competition in generation. The purchasing agency has a monopoly on transmission networks and over sales to the final consumer. This is sometimes called a Single Buyer market structure.

Market Structure 3: Multiple Wholesale Purchasers, which allows choice of supplier for certain sizes of customers, and thus brings competition into both the buying as well as the selling side of generation and wholesale electricity supply. Those who own the distribution wires and retail the electricity are allowed to buy directly from a producer for delivery over a distribution network to their captive customers. This market structure is sometimes called Third Party Access.

Market Structure 4: Retail Competition, which allows all customers to choose their supplier. There is open access to transmission and distribution wires. The natural monopoly activity of distributing electricity to customers using poles and wires is separated from the activity of retailing electrical energy to customers, so that the latter can be competitive.

Figure 3.2.a to Figure 3.2.d show the four market structures schematically. Table 3.2.b shows the advantages and disadvantages of each market structure.


77

Figure 3.2.a Market Structure 1 — Monopoly / the Vertically Integrated Utility

Generator

Wholesaler/Transmitter

Transmission System

DistributionCompany

Consumer

energy sales

energy flows in same company

Source: Competition and Choice in Electricity, Hunt and Shuttleworth 1996

3.2.29 Utilities in Hong Kong are privately owned and vertically integrated (Market Structure 1).


78

Figure 3.2.b Market Structure 2 — Single Wholesale Purchaser

Generator Generator Generator

Wholesale purchasing agent

Distributingcompany

Distributingcompany

Distributingcompany

Customer Customer Customer

energy sales

Source : Competition and Choice in Electricity, Hunt and Shuttleworth 1996

Figure 3.2.c Market Structure 3 — Multiple Wholesale Purchasers

Generator Generator Generator Generator Generator

Transmission Wires

Wholesale Market

Distributioncompanies




Customer Customer Customer Customer

energy sales



79

Figure 3.2.d Market Structure 4 — Retail Competition

Generator Generator Generator Generator Generator

Transmission wires

Wholesale market

Retailer Retailer/DistributionCompany

Retailer/DistributionCompany

Retailer Retailer

Distribution wires

Retail market

Customer Customer Customer Customer Customer

energy sales


3.2.30 In the United States most companies have long been privately held by publicly regulated monopolies (Market Structure 1). However, through the Public Utilities Regulations Policy Act (PURPA), in 1978 competing generators were introduced, but with limited success. (Market Structure 2) Since the Energy Policy Act of 1992, the industry has been in the process of moving slowly to wholesale competition (Market Structure 3) and eventually retail competition(Market Structure 4).

3.2.31 In the UK, restructuring (from a slight variation of Market Structure 1, with multiple distribution companies all state-owned) for full competition started at the same time as privatisation.

Applicability of International Comparisons to Hong Kong

3.2.32 It is worthwhile looking to other countries that have implemented or are implementing electricity sector competition, to see whether Hong Kong can learn from those experiences. However, international experience needs to be applied carefully: only those aspects of the experience that are relevant to Hong Kong should be applied. The discussion below describes the basic pre-reform positions of various countries and considers the important characteristics and experiences from competition in those places. It then considers which aspects of those international experiences may be applicable to Hong Kong.


80

, 03/01/-1,

What does “horizontally and vertically” mean? I think it refers to the deleted graph with the ownership and competition axes, the boxes and the arrows.

Motivations and Factors Driving Change Internationally

3.2.33 The general move internationally from Market Structure 1 to Market Structure 4 has been a result of pressures and issues which have varied from one jurisdiction to another. Major considerations have included: the belief by policy-makers that electricity prices are higher than

they might be in a competitive market; over-investment under rate of return regulation; the need to raise government revenue or to retire government

debt by selling state-owned enterprises, as for example in the UK and the state of Victoria in Australia;

a general political/ideological commitment to private ownership and free market competition, as in the UK, New Zealand and the state of Victoria in Australia leading to the specific view on the part of some governments that electricity supply should be a business (private sector) activity, rather than a government (public sector) activity;

the view on the part of policy-makers that competitive markets offer the best way to limit direct government involvement in the electricity industry;

consumer pressure for competition as a way of reducing prices, as in the United States of America; and

shortage of state funds for investment capital in the electricity supply sector, as in the former Soviet Union and Central and Eastern European countries.

Motivations and Factors that May Lead to Change in Hong Kong

3.2.34 It is clear that some of the reasons for change in other countries do not apply at all to Hong Kong (particularly those related to government-owned utilities) and while others may be relevant only in a limited sense. Specific drivers for change in the electricity sector in Hong Kong include: the general Government policy to promote competition in sectors

where it is feasible; commitments to adhere to the APEC Energy Policy Principles(1)

with a view to the likelihood that at some point in the future, Hong Kong’s electricity market will be integrated with some or part of the South China Pearl River Delta region as noted in Section 1.3.23;

Legislative Council pressure to reduce electricity tariffs; Consumer Council pressure for the promotion of competition to

benefit consumers; and

1(?) APEC Energy Working Group, Adoption and Implementation of the 14 Non-Binding Energy Policy Principles in Hong Kong, China, October 1998.


81

the perception by some that electricity prices are higher than they would otherwise be under competition.

Table 3.2.bAdvantages and Disadvantages of Each Market Structure

Market Structure

Advantages Disadvantages

1 Monopoly / Vertically Integrated

economies of scale; enables the rapid development of

large scale transmission systems; enables the introduction of larger

plants; allows universal service,

development of indigenous fuels and other government policy objectives; and

has the ability to accommodate social policy obligations.

customer takes the risks for mistakes in investment, changes in demand, unanticipated technological obsolescence;

prices may be higher than if under competition.

2 Single Wholesale Purchaser

ability to accommodate social policy objectives;

introduces competition in generation;

avoids some of the costs of later market structures (eg, transaction costs of spot markets and transmission access);

is a good transition market structure where the more sophisticated arrangements needed for a more complete market structure are not in place and would be hard to establish;

provides insurance to the independent generators against market risk and makes it easier to raise capital.

limits the effects of competition, leaving many aspects of the choice of when to build and what to build in the hands of central planners rather than entrepreneurs;

care must be taken to ensure that procurement of fuel and plant operates to ensure least cost outcomes

Comparison between Individual Characteristics of Electricity Sector in Hong Kong in Places that have Introduced Competition

3.2.35 Electricity systems in each country and their reform towards competitive markets may be defined by a number of characteristics, including: the industry market structure (before reform); the form of ownership (before reform); the form of regulation (before reform); the number of generation companies (before reform); the number of individual power stations;


82

, 03/01/-1,

Paul, it would be good to have customer numbers as well as population, and also installed capacity, peak demand and total annual energy consumption. Do you have these figures handy in an almanac to just drop in?

the diversity of generation types; the extent of interconnection between utilities, (in cases

where there is more than one utility before reform); the population of the jurisdiction served by the electricity sector

in question (as a broad proxy indicator of the number of customers in that market);

Table 3.2.bAdvantages and Disadvantages of Each Market Structure (continued)

Market Structure

Advantages Disadvantages

3 Multiple Wholesale Purchasers

expands competition; pushes the market risk and the

technology risk back to the generators.

increases the transaction costs by requiring markets and network agreements;

allocates new technology risk to the generators and also increases their cost of capital;

seen as a transition structure, and is primarily a platform to Market Structure 4;

limited ability to accommodate social policy obligations (eg universal service, reduced rates for low income customers, etc) connected with generation.

4 Retail Competition

for large users, the transaction costs are relatively small per unit of electricity, and sophisticated electronic meters(1) are already in place;

competition should improve both short and long term incentives for efficient production and consumption; and

it can be a pivot in getting rid of many inefficiencies of the previous system

there is no obligation to supply in this market structure, since there are no monopoly franchises;

social policy programmes connected to generation can only be imposed by specific subsidy mechanisms or by an overall sales tax;

it greatly increases transaction costs by requiring more complex trade arrangements and metering of individuals; and

for the small users, the costs may easily outweigh the benefits.

the distribution of customer sizes; the process of reform adopted; and the lessons learned so far from electricity reform.

3.2.36 Some of these characteristics tend to group together. For example, vertically integrated private monopolies tend to be government 1(?) Electricity is a “real-time” commodity, and full energy and financial reconciliation from generation through wholesale purchase and trading to retail consumption across all of the (usually half-hourly) market time periods is a complicated matter.


83

regulated, as in Hong Kong and in most parts of the United States; the reform process for state owned electricity industries almost inevitably involves privatisation. Nevertheless, because the number of characteristics or attributes of electricity systems is relatively long, there are still enough different combinations that each country is unique in some way. For these reasons it is more useful to look at individual characteristics from country to country to see whether any useful comparisons that may be applicable to Hong Kong can be carefully drawn out, rather than simply trying to apply entire experiences from elsewhere. Table 3.2.c provides an overview by individual utility characteristics for comparison and shows which aspects are relevant and which are not relevant to Hong Kong.

3.2.37 The most important point is that electricity sector reform has to be carefully tailored to meet the particular situation and requirements of the place where it is introduced.

3.2.38 Just as electricity sector reform has to be carefully tailored for each particular situation, so the benefits of reform depend on the circumstances in each particular electricity system. In most places, the greatest potential for cost and price savings from competition exists in power generation: generation contributes the largest component of costs, particularly for large consumers. Furthermore, it is usually easier to introduce for competition in generation than in other parts of the electricity sector. An overview comparison table is provided that shows, among other things, that the current potential for generation level competition within Hong Kong itself is limited. Some illustrative examples are provided below of the introduction of competition in places with limited potential for generation competition.


84

Table 3.2.c Comparison of Key Characteristics between the Electricity Sector in Hong Kong and Countries that have Introduced Forms of Competition

Hong Kong England & Wales Northern Ireland

Pre-Reform Market Structure

Vertically integrated, two monopolies

Partially vertically integrated (G&T) monopoly, with several distributors

Vertically integrated monopoly

Ownership before Reform

Two private companies

State owned generation and transmission monopoly

State-owned

Regulation Rate of Return Regulation under Scheme of Control Agreements with closed Government review process

Government oversight of utility management


Number of Generation Companies Prior to Reform

2 1 1

Number of power stations

Limited number (3 major and 1 minor) power stations within Hong Kong; potentially

About 60 Limited number (4)

Diversity of Generation Types

Predominantly coal within Hong Kong; imported electricity from nuclear with pumped storage; increasing use of piped natural gas, plans to use LNG

Coal (decreasing since privatisation and lifting of ban on gas for electricity generation), gas (increasing), and nuclear

Coal and oil

Interconnection between Utilities

End-to-end interconnection between two utilities and large neighbour: currently small HEC-CLP interconnection and larger CLP-GPHC

National transmission grid supplying distribution companies, interconnections to Scotland

Electricity interconnections with the Irish Republic (Eire) were closed at the time of market reform, but have subsequently been re-opened


85

interconnection

Population 6.2 million 51.5 million 1.5 million


86

Scotland United States Australia New Zealand

Vertically integrated, two companies (plus Scottish Nuclear)

Various structures from state to state, including partially and fully vertically integrated

Fully vertically integrated in some states and partial (G&T), with separate distributors in others

Partial vertical integration (G&T), with many small distributors

State-owned Largely private, some state government owned, some municipal, some co-operatives

State Government owned, Local Government owned distributors in one state

State-owned generation and transmission monopoly, many small local government owned distributors


Rate of Return Regulation by State Government utility regulators with open public hearings



3 several hundred SE states: one per state plus shared inter-state hydro, 4 interconnected, plus 2 not yet interconnected

1

Limited number Thousands Over twenty Limited number

Oil, gas, coal, hydro and pumped storage, nuclear.

Varies greatly from state to state and region to region, but includes all major generation types: coal, oil, gas, hydro, nuclear

Varies from state to state, includes black and brown coal, gas, and hydro.

Predominantly hydro which relies on consistent year-round precipitation; some thermal, possible growth in piped natural gas from offshore fields.

End-to-end interconnection between two utilities and large neighbour: Scottish Hydro-Electric to Scottish Power to England’s National Grid; latter interconnection

Most neighbouring utilities interconnected; numerous regional pools operate

Small interconnections between south-eastern states, interconnection planned between Queensland and NSW, long submarine DC interconnection planned between

HVDC interconnection between South Island (where most of the hydro generation is) and the North Island (where most of the load is)


87


strengthened since privatisation

Tasmania and Victoria

5.1 million 269.4 million 18.1 million 3.6 million


88

Table 3.2.c Comparison of Key Characteristics between the Electricity Sector in Hong Kong and Countries that have introduced Forms of Competition (continued)

Hong Kong England & Wales Northern Ireland

Distribution of Customer Sizes

Predominantly small (residential) and medium (commercial), few large (industrial)

Full range from small (residential) to large (industrial)

Predominantly small to medium

Generation Capacity prior to Competition

No significant need for additional new capacity in short term, other than already committed plant, subject to increase in interconnector capacity

No large capacity surplus

Surplus: no new capacity required in short term

Process of Reform

To be decided Separation of generation, transmission and distribution; privatisation, introduction of wholesale market

Single buyer model, generation split into 3 companies and privatised under long PPAs, retail price regulation

Lessons Learned So Far from Electricity Reform

To be decided Small number of large generators able to manipulate and ‘game’ the wholesale spot market

Limited reform can be tailored to small systems

Savings from Competition

Depends on how and when competition is implemented

Fuel cost savings in switch from coal to gas (ban on use of gas for electricity generation removed)

Improved generation plant availability


89


Small and medium, with some large industry



Small (residential) and medium (commercial), with some large (industrial)

Surplus: no new capacity required in short term

Varies from state to state and region to region

Surplus in SE states, no new generation capacity would be needed in short term, if Queensland Government did not resist construction of interconnection with NSW

Severe shortage during drought in early 1990s (strictly speaking a generation energy shortage, or a capacity availability shortage rather than an installed capacity shortage)

Wholesale competition for buyers in UK pool, limited competition between two Scottish generators with transmission constraint, form of price regulation applied, required to take all nuclear output

Varies greatly from state to state; states such as California leading the change, Federal Government attempting to introduce national consistency

Separation of generation, transmission, distribution and retail; privatisation in Victoria, national wholesale market, phased-in retail competition in some states

Separation of generation from transmission; introduction of retail competition between distributors with one state-owned generation company; later split of generation and privatisation

Some form of price regulation is required if there are few generation companies

Early stages Implementation of competition taking longer than originally planned; barriers include labour unions and state governments’ commercial interests

Multitude of tiny distribution businesses did not rationalise as the Government wished and expected, they simply formed retail buying groups

Generation cost reductions have come from ‘yardstick’ regulation

Early stages Large staff reductions prior to privatisation in some utilities; wholesale and retail margins driven down to very low levels; some early buying of market share with negative margins in retail market

Generation level competition is required for significant savings, retail competition alone is inadequate


90

Illustrative Examples

3.2.39 Two examples are provided to illustrate how reform strategies may be adapted where potential for competition in generation is limited. A further example is provided of inappropriate or premature introduction of competition.

Limited Number of Generators – Northern Ireland

3.2.40 Northern Ireland (see Box 3.2.a) has a small electricity system, isolated from the mainland and with only four power plants. Competition of the form introduced in England & Wales would not have been possible and therefore a form of single buyer model was introduced. Subsequently, the interconnections between Northern Ireland and the Irish Republic (Eire) have been re-opened and the market in the Republic has been opened to competition (following the implementation of the EU’s Electricity Directive). Additionally, a sub-sea cable to Scotland will be commissioned. Real wholesale competition has therefore now been introduced in the Province.

Limited Potential for Competition between Generators – Scotland

3.2.41 Scotland (see Box 3.2.b) similarly has a small electricity system whose transmission links to England & Wales were limited. Surplus capacity in Scotland together with an absence of an outside market to sell that capacity, placed a limit on the potential for competition between generators. Two, vertically integrated private companies were therefore created. However, Scotland has more diversity of generation and supply than does Northern Ireland and wholesale competition was considered possible at a much earlier date.

Inappropriate or Premature Introduction of Competition – Ukraine

3.2.42 Ukraine (see Box 3.2.c) provides a different lesson, which may have some application to the consideration of eventual integration of the Hong Kong electricity sector with Mainland China. In Ukraine, full competition is theoretically possible and the England & Wales model was introduced. However, Ukraine continues to have major problems in its transition to a market economy and barter remains an important means of exchange. A competitive electricity market could not hope to operate successfully in these circumstances. On the other hand, it should be noted that though the competitive market is irrelevant, it cannot be accused of causing problems and it does place the Ukraine electricity market in a good position when the economy is reformed.

3.2.43 While the markets in each country are different from that encountered in Hong Kong, there are also many lessons that can be learned from these other countries’ experience. Further examples


91

from the USA, UK, Brazil and the European Union are described in Case Study 2.(1)

Box 3.2.a Northern Ireland’s Experience of Electricity Restructuring

Northern Ireland is part of the UK but is physically separated from mainland Britain. It has a population of 1.5 million and an area of 14 000 km2, the majority of whom live in and around the city of Belfast.Before the restructuring of the electricity industry in the UK, Northern Ireland had a vertically integrated, state-owned power supply company — Northern Ireland Electricity (NIE) — which owned generation, transmission and distribution. The system had a maximum demand of around 1 500 MW and an installed capacity of 2 400 MW including the 800 MW Kilroot dual coal/oil fired power station and a 1 000 MW oil-fired plant which was to be converted additionally to burn natural gas. There were also two relatively old oil-fired plants built during the 1970s.In 1992, the industry was restructured. Unlike England & Wales, where full competition was introduced, a form of single buyer model was introduced. The former electricity utility, NIE was given exclusive responsibility for the purchasing of power from generators. Additionally, NIE, which owns the transmission and distribution businesses and the dispatch centre, was made responsible for system dispatch and for ensuring that it has contracted sufficient capacity to ensure adequate security of supply.NIE was not allowed to own generation assets in the UK. The restructured NIE is obliged to operate separate accounts for its different businesses including:

transmission distribution power purchasing supplyThese businesses are subjected to price regulation based on RPI-X+Y price caps.Until the implementation of the European Union’s Electricity Directive in 1999, the power purchasing business procured power from the generators and sold to NIE’s supply business or to independent supply businesses at the Bulk Supply Tariff (BST). Independent suppliers were permitted to purchase power from NIE’s power purchasing business and sell directly to consumers using NIE’s distribution network. However, the potential for competition was limited to the small margin on the supply business. Few independent suppliers entered this business and those that did, did so in order to establish themselves in the market rather than to make a profit in the short-term.The generation plants which were formerly owned by NIE were separated from NIE and three generation companies were formed and the plants sold to private buyers or to their management. All were sold on the basis of relatively long-term Power Purchase Agreements (PPAs).The PPAs allowed the benefits of efficiency improvements to be retained by the power plant owners. One of the greatest efficiency improvements which the generators have achieved related to availability. Availability payments represent almost exactly half of the cost of the payments to generators (including payment for the cost of the natural gas pipeline). Availability improved significantly since privatisation.The reform arrangements in Northern Ireland were a response to the special

1(?) The material for this section is drawn from a number of sources, namely: Electricity Prices in a Competitive Environment (Marginal Cost Pricing of Generation Services and Financial Status of Electric Utilities) Energy Information Administration, US Department of Energy; Global Utilities Finance Report 1998; and Electricity Liberalisation in Europe, Cameron McKenna, 1997.


92

circumstances in the province; in particular:

insufficient numbers of existing power plants to allow competition in generation;

surplus capacity, so that new capacity was not needed for some time; limited possibilities for trade in a wider international market (though with the

ending of the ‘troubles’ in the Province, exchanges with Eire have resumed).

Box 3.2.b Scotland’s Experience of Electricity Restructuring

Following the restructuring of the UK electricity industry in 1990, the two vertically integrated, state-owned power utilities in Scotland were privatised as Scottish Hydro-Electric and ScottishPower in May 1991.The nuclear assets of Scottish Nuclear were retained in state ownership (but have subsequently also been privatised).Each of the two private companies operates a vertically integrated electrical power system supplying Scotland’s population of 5.1 million with energy sales of approximately 28 TWh per year.Scottish Hydro-Electric’s generating capacity comprises dual oil and gas, hydro and pumped-storage plant while ScottishPower’s capacity is mainly coal-fired with some gas, oil and hydro. Both companies operate fully integrated and interconnected grids. ScottishPower’s transmission system is linked to the national grid in England and Wales by a series of 400 kV and 275 kV interconnectors. The capacity of these circuits was limited to 850 MW at the time of privatisation but has now been upgraded to 2 200 MW. Scottish Hydro-Electric can access the grid in England via the ScottishPower transmission system.Despite being vertically integrated, the two utilities maintain separate accounts for separate businesses. ‘Second tier’ licences authorise the companies to supply customers located outside their supply areas in competition with each other electricity suppliers.Third-party access to the transmission network is mandatory and internal transmission charges are applied to generators and the distribution businesses. Price-cap regulation, of the form RPI - X + Y, applies for transmission, distribution and supply to franchise customers.Wholesale competition is available. However, competition in generation in Scotland is limited by the relatively small size of the market and the large surplus existing capacity relative to demand. The sale of power from Scotland to the market in England and Wales is limited by the capacity of the inter-connector.Because of constraints to competition and allowed cross-subsidies to rural electricity consumers, the generation businesses of the two companies and Scottish Nuclear are regulated. Prices were based on the so-called ‘Great Britain Yardstick’ — the pool price in England & Wales. Adjustments were made in the yardstick to account for differences in the markets in England & Wales in comparison with Scotland.The reform arrangements in Scotland were different from those in England & Wales for several reasons:

the market is small; the potential for trade with the much larger market of England & Wales was

limited by the size of the transmission link; there was substantial surplus capacity and no new generating plant was needed

for some time.


93

Improving the Efficiency of Electricity Utilities

3.2.44 A number of approaches to helping the Hong Kong electricity utilities become more efficient are available, and many lessons have been learned from around the world. This section presents some of the options, and describes how they might be applied to the current market and its possible future variants.

3.2.45 In Hong Kong itself the Government is gaining experience of such market-based control mechanisms. For the last four years the Telecommunications Authority has been introducing measures designed to encourage development of a competitive marketplace in the face of frenetic market activity (see Case Study 5). With regard to electricity, the Government has the opportunity to develop procedures and experience before the current SCAs expire in 2008.

Box 3.2.c Ukraine’s Experience of Electricity Restructuring

Prior to the reform of the Ukraine electricity market the sector was organised into eight regionally based, state-owned vertically integrated monopoly companies. An ambitious reform package was introduced in 1994 which led to the formation of a competitive market consisting of a number of state owned joint stock companies, a Regulatory Commission and Energomarket.The electricity market in Ukraine is large, with total capacity of around 53 000 MW. After restructuring the following were formed: four thermal companies (28 000 MW and 14 power plants), two hydro generating companies (5 000 MW), one nuclear company (14 000 MW in five plants), 27 distribution companies (Oblenergos), a national transmission company, the National Dispatch Centre which operates the competitive wholesale market, Energomarket which is responsible for balancing supply and demand by operating a competitive wholesale market with some similarities to the electricity ‘pool’ in England and Wales.The electricity pool (Energomarket) began operations in mid 1996. Each thermal generation unit which is available must bid a capacity and a price at which it will generate for the next 24 hours.The Wholesale Market Price (WMP) is set by the bid price of the marginal (ie most expensive) of the generators which is needed to be despatched in each hour (the System Marginal Price, SMP) plus some additional costs to cover the capacity payments and administration costs of the market. The WMP is paid to all generators that are dispatched.The market suffers from two main problems. Firstly, large debts from customers have made it impossible to pay generators fully for electricity produced. As a result, the second problem arises. Payments to generators do not cover costs and generators cannot afford to buy fuel. This has meant that almost any generator which has fuel can run — irrespective of least-cost merit order ranking.This experience has shown that it not possible to translate models (in this case the UK pool) to another country without taking account of the specifics of the country involved. Chronic debt problems in Ukraine cannot be solved by such a market. ‘Fuel shortages’ are the result of these debt problems rather than a cause of the problems. The debt problems needed to be dealt with as part of the restructuring.It is important to note, however, that the problems which exist at present are not the result of those reforms. Non-payment by customers, non-payment for fuel, fuel shortages, poor maintenance practices and low availability would all have occurred in the absence of the competitive wholesale market. The reform may have been


94

unnecessarily complicated at this stage of transition to a market economy, but the reform has not worsened the problems. Also, in its favour, it should be noted that Ukraine now has the foundations of a properly functioning electricity market in the future — when the right conditions emerge.The real criticism of the competitive wholesale market in Ukraine is not that it caused problems, but rather that it was an unnecessary waste of resources.

3.3 Evaluation of Four Market Structures Applied to Hong Kong

3.3.1 This section considers the four various market structures identified above in the context of the development of Hong Kong’s electricity sector and ultimately presents possible models for consideration. The subsequent section then evaluates the likely costs and benefits to Hong Kong consumers of developing the market according to four chosen scenarios.

Characteristics of Hong Kong

3.3.2 There are currently a number of limitations to the introduction of competition within the generation market in Hong Kong. Limitations that are characteristic of the existing situation include: only two existing electricity companies in Hong Kong; a small number of generators (three major and one minor power

station within Hong Kong itself, plus some contracted supply from Mainland China);

limited potential for generation fuel competition; and no major industrial customers.

3.3.3 Limitations on generation-level competition — including limitations on the introduction of new generation companies in Hong Kong — are detailed below in Table 3.3.a.

Options For Hong Kong Within Each Market Structure

3.3.4 The rest of this Section therefore provides: a discussion of the four market structures as they might be

applied in Hong Kong; the structural issues encountered in Hong Kong; the alternative market structures considered most appropriate for

Hong Kong; and alternative forms of regulation that could be applied to the parts

of the electricity supply industry that are not open to competition.


95

Market Structure 1 — Vertically Integrated

3.3.5 As mentioned previously, the actual situation in Hong Kong corresponds closely to Market Structure 1. As the implementation of a vertically integrated utility structure is probably the most straightforward market to implement, it has fitted well with the Government’s general philosophy of minimal intervention in markets.

Table 3.3.a Potential Limitations on Competition in Generation in Hong Kong

Issue DiscussionLimited and extremely expensive land

Land area is extremely limited in Hong Kong, and although additional land can be created, reclaimed land is also very expensive. Government subsequently has a number of options if it wants to encourage additional competition in Hong Kong: designate more land for additional generators; require the two utilities to tender for additional capacity

and allow the tenderer to build on the sites that the current utilities intend to utilise (eg future extension on reclaimed land at HEC’s Lamma Island site; remaining land at CLP’s Black Point site and/or land from future additional cut-and-fill at that site); and

allow IPPs to bid on plants to be built in Mainland China, or to utilise the Southern Chinese electricity grid, hence avoiding the issue of additional land requirements in Hong Kong.

Potential lack of legal ability to split generation assets from the existing utility ownership to create more competitors

One of the options to introduce further competition would be to split generation capacity from the existing utilities and create new companies. The following should be noted: New legislation is likely to be required to force the utilities

to undertake this option. Within Hong Kong there would only be three main

producers (Lamma, Castle Peak and Black Point) and potentially one small one (Penny’s Bay).

The contracts with Daya Bay and Guangdong Pumped Storage would have to be honoured.

Interconnection Capacity

Competition in electricity generation requires the competing parties to be interconnected. The existing utility systems in Hong Kong are interconnected with one another and one is interconnected with the Southern China system. The transmission capacity between supply (generation units) and demand (customer loads) constrains the net flow of energy between those parties which may constrain transactions between those parties. It is highly likely that the HEC-CLP interconnection capacity would need to be increased to make electricity sector competition in Hong Kong feasible and to realise any potential benefits from competition.


96

Issue DiscussionPotential for exercise of significant market power

This consideration arises in addition to the general observations on interconnection capacity. In situations where land is limited and new plants cannot easily be built, market power might be exercised by the incumbent. This would arise in particular with isolated systems where interconnection capacity is limited. This situation would exist on Hong Kong Island in the absence of increased interconnection capacity to Hong Kong Island (See the last section of Case Study 3 for further detail on potential market power on Hong Kong Island).

Lack of market development in Southern China

A fully developed, competitive electricity supply market in Southern China would help provide a relatively easy way for Hong Kong to introduce competition. However, the market has to develop significantly before Hong Kong could feel comfortable in acquiring power through a competitive Southern China market.

3.3.6 However, the Consultants view, supported by international experience, is that the effectiveness of this arrangement and benefits to consumers could be improved by financial or strict accounting separation of the businesses into the separate functions of generation, transmission, distribution and retail customer services. As shown in Section 2, consumer benefits could also be realised from increasing the interconnection capacity and operating a Firm Capacity Only or Firm Capacity and Firm Power pool. Such pooling arrangements, though, would not in themselves promote competition.(1) Hong Kong may be considered to have an informal electricity price cap as a result of the utilities’ desire to avoid public criticism associated with tariff increases. This has encouraged operating efficiency to be maintained at good levels or improved and tariff increases have remained below consumer price inflation in recent years. Nevertheless, the incentive to over-invest that is inherent in rate-of-return regulation remains.

Market Structure 2 — Single Wholesale Purchaser

3.3.7 In summary, Market Structure 2 in Hong Kong would require that all new generating capacity be developed by independent companies (possibly including companies associated with HEC and CLP — but with separate accounts). The need for new capacity would be announced and tenders invited. The tenders would be evaluated by an independent body according to defined and transparent rules. The company winning the tender would enter into a long-term power purchase agreement with the single wholesale purchaser and would be dispatched by the ISO on the basis of that contract.

3.3.8 Under this structure, downward pressure would be exerted on the costs of new generation but not necessarily requiring major

1(?) See paragraph 3.2.34 for a discussion on some of the drivers for change within the Hong Kong electric utility industry.


97

legislative or structural changes. The main gains would be achieved by removing new generation assets from the Scheme of Control and allowing the rate of return to find its own market level.

3.3.9 As an initial step towards competition, these Single Wholesale Purchaser systems could be achieved within the two present utility territories under any scheme of co-operation (eg, emergency support, Firm Capacity Only, Firm Capacity and Firm Power — the choice of scheme would not influence fundamental economic behaviour).

3.3.10 Within almost any variant of Market Structure 2, it would be possible to oblige utilities to purchase the output from cogeneration, small renewable plants and waste-to-energy facilities at regulated prices, as occurred with the introduction of the PURPA(1) legislation in the USA. This would help the Government to achieve environmental objectives, bringing further benefits to consumers. Over time, this would move a major part of the utilities’ current capital expenditure outside the SCA and ensure market rates of return. It could also reduce the incentive to over-invest in plant.

Market Structure 3 — Multiple Wholesale Purchasers

3.3.11 In the context of Hong Kong, the Multiple Wholesale Purchasers market structure could begin by having large consumers buy from either power company.

3.3.12 A scenario that is more likely to be successful would be to enable large consumers in Hong Kong to buy from independent power producers or suppliers in Mainland China, a step considered crucial to the success of Multiple Wholesale Purchasers in Hong Kong. This step would most likely need to be combined with electricity competition legislation to require the utilities to permit third-party access to their transmission and distribution networks on non-discriminatory terms.

3.3.13 The opening of the market could be controlled by specifying dates for progressive levels of market opening in terms of the characteristics of the consumers that would be eligible to participate. Appropriate regulatory tools would also be required to prevent abuse of market power by the dominant players. See Case Study 5 for examples of how the telecommunications market was opened up in Hong Kong and how checks were put in place to prevent dominant market behaviour.

3.3.14 Determining who should have the ultimate ownership of the transmission system in Hong Kong is more difficult again than the issues relating to generation or distribution. It should be noted that

1(?) The Public Utilities and Regulatory Policy Act required utilities to set prices at which they would purchase power from non-utility generators. The success of PURPA has been debated, but many consider it to have laid the foundation for further competition as it facilitated the development of an independent power producer (IPP) industry.


98

this is a different issue from the technical operation of the transmission system.

3.3.15 As long as the transmission system is managed in such a way as to ensure that access is on a non-discriminatory basis, it may be possible to maintain ownership of the assets as they currently exist and still implement an independent transmission operator, or ISO. However, separate ownership greatly simplifies the issue as fewer concerns about abuse of market power are likely to arise if ownership is separated from generation and distribution.

Market Structure 4 — Retail Competition

3.3.16 Under Retail Competition all classes of consumers — from large industrial plants to the smallest domestic customer — can exercise choice in who their supplier may be, such as the generator, distributor, transmitter or retailer (eg, power marketer). In Hong Kong, there is a precedent for Retail Competition in the telecommunications industry. However, the telecommunications industry is a much easier industry to deregulate as it is essentially a network, without the significant fixed costs of generation plant. It is also considered that once Multiple Wholesale Purchasers begins, it is likely that the end result will be full Wholesale and Retail Competition as one customer may perceive gains from participating in a market from which they may be excluded. Hence Government would likely reduce the threshold for participation (eg, amount of energy consumed) over time in response to provide competitive market choice to more customers, leading eventually to full Wholesale and Retail Competition.

Alternative Forms of Regulation for the Non-Competitive Parts of the Industry

3.3.17 The electricity supply industry is usually thought of as being made up of a number of functional parts: generation, transmission, distribution and energy retailing. It is possible to introduce competitive markets in (wholesale) generation and in electricity retailing, but it is not considered practical to introduce direct competition in transmission and distribution functions. Those network businesses are generally considered to be “natural monopolies” because it tends not to make economic sense to duplicate them. So even in places where competition has been introduced, local transmission and distribution monopolies remain under some form of regulation.

3.3.18 Enough experience exists to show that retail competition for large and medium-sized (industrial and commercial) customers is a practical proposition. However, it is not yet clear from new competitive electricity markets around the world whether or not it will be viable to implement retail competition right down to the level of small residential consumers, although this is in the early stages of


99

implementation or soon to be introduced in some jurisdictions, including New Zealand, England, some parts of the United States and at least one state in Australia. Consideration has consequently been given in this Study to alternative forms of price regulation. Where competition in retail electricity supply is not available to some or all customers, some form of regulation is required.

3.3.19 Means of improving utility efficiency may therefore be divided into two parts, one considering likely utility performance under alternative forms of regulation applied to those parts of the electricity sector that are not subject to competitive market pressures and the other considering likely utility performance under those parts of the electricity sector that are subject to the operation of competitive markets.

3.3.20 It is generally accepted that the prices charged by monopoly businesses need to be regulated. This is implicitly recognised in Hong Kong’s Scheme of Control Agreements between Government and each of the electricity companies. There are three main approaches to price regulation: rate-of-return regulation, which is the methodology currently

practised in Hong Kong; yard-stick regulation; and performance-based regulation using a price-cap or a revenue

cap.

Rate of Return Regulation

3.3.21 Rate-of-return regulation of private monopolies, has historically been applied widely, particularly in the United States electricity industry. This form of regulation is the basis for the SCAs that govern the utilities in Hong Kong. These agreements have been very successful in the past at ensuring that the utilities satisfy the rapidly growing demand for power in Hong Kong and providing a high level of reliability while placing both a ceiling and a floor on the electricity companies’ profit margins. It may nevertheless be timely to review the type of regulation utilised in Hong Kong in an effort to establish whether it is still the most appropriate methodology.

3.3.22 General strengths of rate-of-return regulation include the investment certainty that it provides; and the obligation to supply is usually placed on the regulated company.

3.3.23 General weaknesses of this form of regulation include: the lack of incentives for utilities to reduce capital expenditure /

the perverse incentive for the utility to increase capital expenditure because total profits are a function of the permitted return and the value of the asset base (the permitted return must be greater than the cost of capital to provide the utilities with an incentive to invest at all; and


100

the need for Government, or a regulatory body, to scrutinise each investment to protect consumer interests, which tends to duplicate to some extent the functions of management and can become expensive.

3.3.24 Under this form of regulation, the perverse incentive to over capitalise, if not effectively checked can lead to high reserve margins for generation; over-specification of plant; little control of procurement costs; preference for capital intensive technologies (eg coal over combined cycle gas turbines); and incentives to excessively reinforce networks. As noted above, the SCAs in Hong Kong have recently been amended to provide a mechanism to discourage investment in excess capacity.

Yard-Stick Regulation

3.3.25 Yard-stick comparisons seek to ensure that the regulated entity is in line with similar entities in other places. Yard-stick comparisons may provide useful indicators of general economic efficiency levels, but in practice all utilities have differences with respect to technology, accounting practices, market structure, load density and age of plant, which make direct comparison difficult. Adjustment can be made for each of these factors, but each adjustment adds a degree of uncertainty and reduces confidence in the conclusions. Yard-stick comparisons are useful, but are not a complete basis for regulation.

3.3.26 Proposals for yard-stick regulation were submitted to the Federal Energy Regulatory Commission in the US covering the interstate natural gas pipeline industry. The proposed arrangement gave the pipeline operators a cost-based price but future indexing of prices was based on industry-wide changes in costs and productivity. Yard-stick competition would, in this instance, have been possible because the US has a large number of gas pipeline owners/operators (21 of them) whose costs have similar drivers. The scheme was not, however, implemented. An arrangement similar to this would be difficult in Hong Kong because of the lack of utilities with comparable characteristics and comparable cost drivers.

Performance Based Regulation

3.3.27 Performance-Based Regulation (PBR), using either a price-cap or a revenue-cap approach, aims to set a ceiling on the revenue of the utility, either directly or as a function of measurable operating parameters (eg units sold, customers connected). Price caps may be set with a time-based adjustment to guarantee a portion of benefits to consumers over the inter-review period. The cap is stable for a period, typically five years, after which it is reviewed. The utility therefore has an incentive to reduce costs, because it can keep the savings until the next price review.


101

3.3.28 Price-caps tend to work better when they are applied to the separate business functions of a utility, ie transmission and distribution and possibly supply and generation, because the capping formula can be adjusted to provide incentives appropriate to each business function.

3.3.29 Price-caps are not a perfect solution. The regulated business will still calculate whether the immediate benefit of a cost reduction exceeds the loss from a subsequent downward revision of the price-cap and the evidence shows that eventually price-capping deteriorates into a form of rate-of-return regulation, because when prices are reviewed reference is made to an acceptable rate-of-return on assets.

Most Appropriate Options for Hong Kong

3.3.30 Any process of change in the electricity sector would take some time. Therefore, in the immediate future, the existing electricity market structure will continue in Hong Kong. Nevertheless, some opportunities for capturing economic and consumer benefits exist in the areas of: increasing the capacity of the interconnection between HEC and

CLP as described in detail in Section 2; and increasing use of the interconnection between HEC and CLP as

described in detail in Section 2; making some adjustments to the form of regulation, either in the

final SCA review in 2003 or for the period after 2008.

3.3.31 In the short term to 2008, Government may consider some modifications to the existing arrangements for the vertically integrated electricity sector (Market Structure 1) appropriate, particularly to capture the benefits of increased interconnection described in Section 2. This would require adjustments to the SCAs at the review for the final five-year phase from 2003 to 2008 of the current 15 year SCAs, as well as re-negotiation of the ICA between the utilities.

3.3.32 In the medium and longer term, beyond 2008, Market Structure 2 — Single Wholesale Purchaser and Market Structure 3 — Multiple Wholesale Purchasers would be the most appropriate for consideration in Hong Kong. These are assessed in detail in the next Section.


102

3.4 Detailed Consideration of Options Appropriate to Hong Kong

Criteria for Assessment of Options

3.4.1 A set of criteria has been selected to assess which of the options for alternative electricity sector market structures are most appropriate for Hong Kong. These criteria may be grouped under three headings: that the system operation should ensure inexpensive and

reliable power; that new facilities should involve least cost procurement; and that the system should allow a degree of regulatory

oversight appropriate to ensure reasonable overall economic results and protect consumer interests without the need for Government to intervene excessively nor to incur high costs of regulation.

Criteria for System Operation

3.4.2 A power system should operate its plants in the most economic fashion. It should price its output in a manner that reflects the costs of supply at different times of the day to different groups of consumers (ie there should be no abuse of market power). Criteria for good performance of the system include: incentives to reduce costs of procuring fuel and of operating

plant; incentives for efficient utilisation of investment (high

availability of plant and economic dispatch); that the system should operate reliably and not be susceptible to

the failure of some players (no risk of systemic failure); mechanisms to ensure cost-reflective tariffs; and mechanisms to ensure that there is no abuse of market

power.

Criteria for Investment in New Capacity

3.4.3 The electricity market must be capable of ensuring adequate generating capacity in the future at the lowest cost consistent with the necessary reliability and quality of supply. Investments should also be commercially robust in that they choose appropriate fuels and technology. The performance criteria in this respect can be specified as: incentives for adequate and effective investment; and incentives for procurement at least cost.


103

Future Regulation and the Role of Government

3.4.4 Government will inevitably have a role in controlling and regulating the market for electricity. Its functions may include: price regulation of monopoly elements of the industry; licensing of participants and potentially of construction of plant

and equipment; supervision and control to prevent abuse of market

power; and regulation of safety aspects, in relation to both workers and

also the general public, environmental impacts and land-use issues.

3.4.5 Some structures may pose different challenges to government, for example, a competitive sector may resist investment in environmental control equipment more strongly than a vertically integrated industry, due to the pass through of costs under a Rate of Return regulation.

3.4.6 Performance criteria relating to the role of government and regulation include: that the structure should be capable of effective price

regulation where necessary; that there are no problematic technical issues concerning

safety, land-use and the environment; that the resources required for regulation should be

acceptable; and that the role of government should be minimised.

Market Structure 2: The Single Wholesale Purchaser

3.4.7 There are three possible variants of Market Structure 2: a Single Wholesale Purchaser, purchasing from qualified

producers (renewables, cogeneration, waste plant etc); a Single Wholesale Purchaser with some existing plants replaced

by IPPs; and a Single Wholesale Purchaser with an obligation that all new plant

should be constructed as IPPs. These are each assessed against the criteria specified above.

Operation of the System

3.4.8 An early stage in the development of IPPs internationally was the obligation introduced in the USA for utilities to purchase power from ‘qualified producers’. These producers were sources which were perceived by the Government and regulators to be economically and


104

, 03/01/-1,

What does Paul mean by “existing” plants? Should “replaced” be changed to “displaced”. In other words, the existing plants remain, but are in competition with this new capacity. Is this some form of over-capacity to promote competition? Or does it mean that existing plants are decommissioned to make way for IPPs? Or does it mean that some existing plants are compulsorily sold to IPPs to enable a competitive market?

environmentally beneficial, but which were prevented from entering the market by the market power of incumbent generators. The PURPA legislation obliged utilities to purchase power from these utilities at a cost based on a centrally determined assessment of what is known as ‘system avoided cost’ . Such legislation creates the simplest form of Single Wholesale Purchaser.

3.4.9 The experience in the USA was not entirely satisfactory, because it turned out that avoided system costs subsequently fell, so purchasing obligations under long-term contracts caused costs and consumer prices to be higher than they otherwise might have been.(1)

3.4.10 Hence, whether or not to offer protection to qualified producers is a decision that should be taken separately from the decision about market structure. All market structures can be adapted to subsidise specific forms of production if that is thought desirable, although the mechanisms will differ. The first variant of Market Structure 2 considered in this Study is therefore altered as it can be subsumed within a number of other structures.

3.4.11 Implementation of the other variants of the Single Wholesale Purchaser market structure will depend in each variation on contracts that are agreed between the generator and the Single Wholesale Purchaser, and the fuel provider and the generator. These contracts should be designed to maximise the utilisation of generation plant, to ensure economic dispatch, to allocate economic risk efficiently and to put downward pressure on costs. Contracts can be designed to achieve these objectives. Given that the contracts between power plant and the Single Wholesale Purchaser/s would be the main determinant of behaviour and that competition for new load would be the main source of competition, it may not be a prime consideration whether to require a change in the ownership of some existing plant.

Provision of New Facilities

3.4.12 New plant in each of the Single Wholesale Purchaser variants would be procured by competitive tender, so there is no distinction between the variants against this criterion. For the same reason, all of the variants would provide the same incentives with respect to adequate and effective investment. The advantage over the current arrangements would be that plants independent of the existing utilities would be likely to offer lower overall costs, including lower rates of return than current arrangements provide.


3.4.13 Each of the variants would have similar implications for future regulation and the role of government, since price regulation would

1(?) Refer to Case Study 4 on Stranded Costs..


105

follow the same lines in each of the market structures. The resources required would not differ significantly and because new plant is procured in the same way in the variants there would be no differences in terms of land-use and environmental impact. Arrangements would have to be introduced for planning and tendering of new plant by an independent body such as the appropriate Government regulator.

Market Structure 3: Multiple Wholesale Purchasers

3.4.14 There are three possible variants of Market Structure 3: Multiple Wholesale Purchasers for large medium consumers with

one ISO; as above, but with two ISOs; and creation of independent distribution companies with rights of

Multiple Wholesale Purchasers.

3.4.15 The variants proposed in this market structure reflect the question of whether to have one or two transmission system operators and which customers should be eligible to buy from generators either directly or through a third party independent supplier.

3.4.16 International experience of introducing Multiple Wholesale Purchasers has often proceeded progressively from large industrial consumers to medium and then small consumers, and finally to retail access. In some countries, especially where there are municipally-owned distribution companies, these companies have been made eligible consumers.

3.4.17 Because Hong Kong currently has vertically integrated electricity companies, separate distribution companies do not exist. Distribution companies could theoretically be created by breaking up the distribution system around existing geographical areas.

3.4.18 The variants for consideration are discussed in the following sections:

Operation of the System — One or Two Independent System Operators?

3.4.19 Considering first the manner in which the system would operate with either one or two ISOs.

3.4.20 The main requirement for Multiple Wholesale Purchasers is that the size of the interconnection should be such that competition for eligible consumers is not hindered by constraints on interconnection and this condition will need to be met regardless of the number of ISOs. It is not obvious at this point how large that interconnection would need to be to satisfy that requirement and it depends to some extent on the size of the competitive market.


106

3.4.21 From the point of view of maximising the gain of new customers outside their historical supply areas, a strong interconnection would be important to CLP and essential for HEC. From the point of view of minimising the loss of existing customers, a weak interconnection would be important to CLP and essential for HEC.

3.4.22 The second condition is that IPPs should have non-discriminatory access to the transmission network. This can be assured by a single ISO. It is more problematic if there are two ISOs, because the procedure for access to the transmission network may differ between them.

3.4.23 If these two conditions hold, then the captive market in each area could be served by the existing utility structure, assuming financial and managerial separation of transmission from generation and distribution. But, at this point the argument for pooling applies and the logic of joint dispatch and planning for the captive market is again valid and the market structure for a single ISO follows.

3.4.24 We conclude therefore that Multiple Wholesale Purchasers could be organised with separate ISOs. It would be considerably simpler to work with a single ISO and it is likely to lead to lower costs through better dispatch.

3.4.25 The other operational criteria do not differentiate strongly between one or two ISOs. By optimising the use of assets, a single ISO could reduce the risk of systemic failure.

Operation of the System — Definition of Eligible Consumers

3.4.26 In general terms, the more consumers that are eligible to participate in the marketplace, the stronger the competitive pressures. Theory and international experience both suggest that competition reduces procurement costs and leads to more efficient operation. However, it may also increase the cost of capital, because of the additional risk to investments. International experience to date suggests that on balance costs and prices fall.

3.4.27 These effects would be enhanced by a greater degree of market opening and therefore the market structure variants which offer greater access and larger market openings are to be preferred according to all these criteria. Cost-reflective tariffs would apply to more consumers, market power of generators would be reduced and competitive pressures to reduce costs would apply more widely.

3.4.28 Hong Kong does not have many large industrial consumers and it is questionable whether they would be sufficient to support a reasonable market opening, as the Consultants estimate that a market with 20-30% of total sales is required to have effective competition. HEC’s industrial sales are approximately 5% of its total and CLP’s are approximately 15%.


107

3.4.29 However, Hong Kong does possess many large buildings with significant electrical load. It would be technically and commercially feasible for supply to these buildings to be permitted under Multiple Wholesale Purchasers. This could be the most likely group of eligible consumers. This market structure would correspond to implementing Multiple Wholesale Purchasers for both large and medium consumers simultaneously, rather than in two or three steps, as has been done in places with many large customers.

3.4.30 Theoretically, opening the market could also be achieved by creating separate distribution companies. These companies could then be eligible wholesale purchasers in their own right alongside, or as an alternative to, medium sized individual consumers. (This variant is discarded later in the Study on the grounds of the very high transaction costs that would be incurred, so its performance on the basis of the operational criteria is not analysed).

Provision of New Facilities

3.4.31 This criteria is not relevant to the choice between one and two ISOs.


3.4.32 Effective implementation of Multiple Wholesale Purchasers does require that there are generators willing to compete for customers defined as eligible to purchase from any supplier. It serves no purpose to designate such a group of consumers if no generators are willing to enter contracts to sell to them directly.

3.4.33 In Hong Kong it is possible that collusion among the incumbents might prevent new generators from entering the competitive market and new generators might be deterred from entry by the prospect of predatory pricing(1) by the incumbents. This may particularly be the case when there are relatively few large consumers and the new entrant is faced with the need to build a portfolio of medium-sized plants before its investment can be repaid.(2) A solution to this dilemma is to create a Single Wholesale Purchaser system in which the generators are divested from the utilities. Each generator is initially contracted to the Single Wholesale Purchaser but has free capacity that it can sell either to the Single Wholesale Purchaser or to eligible consumers. Providing there is initially some surplus of

1(?) Predatory pricing behaviour involves selling at a loss for a period of time to force a competitor out of business. Larger companies, well established in their market are often in a position to use this business strategy against new entrants with smaller operations in that market. New entrants with a large business base elsewhere may be able to withstand such a strategy, but may not consider it worthwhile to risk the possibility of being drawn into a price war and incur losses soon after entering a new market. In this way, the perception on the part of potential new entrants of the threat of predatory pricing by incumbents may be sufficient to deter them from entering that market. 2(?) Examples exist in markets with two suppliers of collusive or at least simultaneous predatory pricing to force out a third, newly entered competing supplier. This happened relatively recently in the domestic airline business in Australia, where Qantas (domestic) and Ansett were able to drive out the new entrant Compass by continually meeting or undercutting its lower prices until it went broke. Prices then returned to higher levels.


108

capacity and providing there are enough power plants this arrangement will create effective competition.

3.4.34 Given that there are only a handful of power plants in Hong Kong at present and that this will not change significantly in the near future, it seems likely that the only way in which Multiple Wholesale Purchasers could be established in Hong Kong would be when the Southern China market matures to become a real source of competition.

3.4.35 Implicitly, it has been assumed that the Multiple Wholesale Purchasers model comprises a continuation in a modified form of one or two vertically integrated utilities but that a share of the market is opened to competition from independent generators or suppliers making use of the transmission system via an arrangement known as third-party access. A variant of this would be to separate distribution from transmission and generation.

3.4.36 In restructured electricity markets, ‘distribution’ is thought of as comprising two (or three) distinct activities. These are “distribution” itself which is the ownership and operation of the physical distribution network; (1) “supply” which is the buying and selling of electricity; and “metering” although metering is often included as part of one or other activity. In some electricity markets, the activities of the distribution companies are treated as separate businesses with separate accounts. There are possibilities for competition in supply and the “supply” businesses in markets with full retail competition (as has become the case recently in the UK) may be unregulated. Competition is not possible for the distribution or “wires” business.(2)

3.4.37 In a multiple wholesale purchaser market structure with independent distribution companies, the distribution companies could also be free to purchase power from the lowest cost supplier. This would add further to downward cost pressures on generation. The choice between one, two or more distribution companies will have little bearing on competitive pressure on generation costs. Alongside the distribution companies, medium and large customers would also be free to purchase power from the cheapest source at unregulated prices. The captive customers would be obliged to purchase their power from the distribution company so that the supply ‘business’ of the distribution company would need to be regulated.

3.4.38 This variant remains a possibility but would require further detailed investigation when and if Government chooses to implement market structure 3.

1(?) Readers may have seen other authors call this the “poles and wires” business, or simply the “wires” business. 2(?) This is because it is considered a natural monopoly within a given geographical area. In other words, it does not make economic sense to duplicate the network because the savings induced by competition would be more than offset by the cost of duplicating the network.


109

3.4.39 It is theoretically possible to have distribution companies in Hong Kong, separate from transmission and generation, who would buy electricity in a competitive wholesale market. However, for the necessary competitive market forces to exist there could not be common ownership between generation (wholesale sellers) and distribution (wholesale buyers). Therefore, if this approach were to be adopted, Government intervention would be required to break up the distribution function of the existing companies into separate businesses.

3.4.40 However, the costs of breaking up the distribution companies would tend to be high relative to the benefits to be gained. It would also likely be a very difficult task from both a legal and a policy point of view, requiring significant Government time and effort. Such effort would need to be very carefully assessed against the benefits, and compared with the alternative options before embarking on such as course. The Consultants do not consider that this possibility would be a sensible option given the existing situation in Hong Kong.

3.4.41 So, to provide a sufficient number of wholesale purchasers to achieve a significant market opening, it would be necessary to permit medium-sized consumers such as large offices and apartment blocks to be eligible consumers; a direct jump to Multiple Wholesale Purchasers for medium consumers would therefore be appropriate under Market Structure 3. But until the development of the Southern China market allows it to be opened up to integration with Hong Kong, the creation, or introduction, of multiple generators seems unlikely and this therefore precludes this structure in the immediate future.

Most Appropriate Forms of Each Model for Hong Kong

3.4.42 The market structures as they apply in Hong Kong are discussed below and schematic representations are included in Figure 3.4.a to Figure 3.4.c.

Market Structure 1 — Changes under the Existing Structure

3.4.43 The changes under the existing structure relate to the implications for the existing institutional arrangements (including the SCA and the IA) of each of the interconnection / generation expansion scenarios analysed in Section 2.

Market Structure 2 — Single Wholesale Purchaser

3.4.44 The proposed form of Market Structure 2 is the Single Wholesale Purchaser model where the existing generating assets remain with the existing companies but new generation is procured under competitive tender managed by an independent third party such as a Government regulator.


110

3.4.45 This proposed market structure would require the creation of a body with responsibility for evaluating tenders for IPPs; the subsequent tendering of the IPP contracts.

3.4.46 The current SCAs each contain a clause about the companies obligation to meet future demand for electricity:

3.4.47 China Light recognises its continuing obligation to contribute to the development of Hong Kong by providing sufficient facilities to meet the present and future demand for electricity and in pursuit of this objective, with the approval of the Government, has entered into agreements with [third party companies] under which [joint-venture company] would construct additional generating capacity for sale of electricity to China Light. (1)

3.4.48 HEC recognises its continuing obligation to contribute to the development of Hong Kong by providing sufficient facilities to meet future demand for electricity and in pursuit of this objective, would construct additional generating, transmission and distribution facilities for sale of electricity to its consumers. (2)

3.4.49 These clauses are a form of what is known in the electricity industry as the “obligation to supply.” In many jurisdictions, this obligation has been required of private monopoly utilities as a condition of granting monopoly franchise. However, Government has made clear to the Consultants its view that in Hong Kong, neither the SCAs in general, nor the ‘obligation to supply’ clause in particular, grant the existing companies exclusive monopoly franchise over the areas that they serve.

3.4.50 On this understanding of the SCAs, competitive tendering of new generation by an independent party (Government) would not require major changes to the current SCAs. The Consultants note that the two utilities may not necessarily agree with this understanding of the SCAs and may challenge such an interpretation.

1(?) The Scheme of Control Agreement entered into by the Hong Kong Government and the following Companies: CLP Hong Kong Limited, Exxon Energy Ltd. and Castle Peak Power Company Ltd, Clause (B).2(?) The Scheme of Control Agreement entered into by the Hong Kong Government and the following Companies: the Hongkong Electric Company, Limited and Hongkong Electric Holdings, Limited, Clause (D).


111

Figure 3.4.a Market Structure 1 in Hong Kong — Two Vertically Integrated Monopolies

HECHEC CLPCLPGeneration

System Operation

Distribution,RetailSupply

HECHEC CLPCLP

HECHEC CLPCLP

TransmissionOwnership HECHEC CLPCLP

Figure 3.4.b Market Structure 2 in Hong Kong — Single Wholesale Purchaser


System Operation


ISO

HECHEC CLPCLP

TransmissionOwnership HECHEC CLPCLP

CLPMainland

IPPs

Market Structure 3 — Multiple Wholesale Purchasers

3.4.51 The proposed Market Structure 3 variant is Multiple Wholesale Purchasers for medium sized consumers with a Single Wholesale Purchaser for captive customers (eg too small to qualify for Multiple Wholesale Purchasers) and a single ISO.


112

Figure 3.4.c Market Structure 3 in Hong Kong — Multiple Wholesale Purchasers


System Operation


ISO

TransmissionOwnership HEC, CLP, Other?HEC, CLP, Other?

MainlandIPPs

MainlandIPPs

HECHEC CLPCLP LargeCustomers

LargeCustomers

3.4.52 This proposed market structure would require: a developed market of IPPs, with the exact number yet to be

determined; most likely, more extensive integration of the Hong Kong

electricity grid with the Southern China electricity grid, as it is unlikely that a sufficient number of IPPs can be developed in Hong Kong;

expiration or revocation of the existing SCA agreements; the introduction of third-party access to the transmission (and

distribution) networks; negotiation with utilities on the ownership of transmission assets

and the operation of those assets (eg the single ISO); definition of ’medium-sized customers’; and an approach to the regulation of the transmission and distribution

networks.

Market Structure 4 — Retail Competition

3.4.53 Market Structure 4 is Retail Competition, and is viewed as the inevitable extension of Market Structure 3. In addition to the requirements of the proposed Market Structure 3, Market Structure 4 would require creation of supply organisations aimed at small retail customers.


113

Market Transition

Possible Trajectories of Market Structure Transition

3.4.54 This section provides some thoughts on the timing of possible trajectories of market structure transition if opening the market to competition is to be pursued.

3.4.55 There are four general market structures. As discussed previously, Multiple Wholesale Purchasers (Market Structure 3) is not considered to be a permanent option in the long run, since the natural progression would be to develop it into Retail Competition. Consequently, market transition trajectories that end in Market Structure 3 (Multiple Wholesale Purchasers)are not shown, leaving five market transition trajectories in addition to the status quo, as shown in Table 3.4.a.

Table 3.4.a Market Transition Trajectories in Hong Kong

Market Trajectories

Status Quo: Structure 1Trajectory 1: Structure 1 Structure 2Trajectory 2: Structure 1 Structure 2 Structure 3 Structure 4Trajectory 3: Structure 1 Structure 3 Structure 4Trajectory 4: Structure 1 Structure 2 Structure 4Trajectory 5: Structure 1 Structure 4

3.4.56 Given: the relatively complex array of considerations involved at the

detailed level of planning and actually implementing competition in electricity markets, and

the fact that a competitive electricity market in Hong Kong would in its particular combination of attributes be unique,

the consultants consider that it would not be advisable to introduce full Wholesale and Retail Competition (Market Structure 4) in one step, ie without initially implementing some form of wholesale competition (Market Structure 2 or 3) and first resolving any particular problems encountered at that level of the market. For this reason, trajectory 5 is not considered advisable.

3.4.57 Furthermore, the introduction of Multiple Wholesale Purchasers would likely provide a smoother transition to full Wholesale and Retail Competition than would moving directly from Single Wholesale Purchaser to full Wholesale and Retail Competition. Therefore, trajectory 4 is not considered advisable.

3.4.58 Trajectory 1 would be chosen by Government if it decides on limited reform, but decides not to move to the introduction of full competitive markets.


114

3.4.59 Trajectory 2 represents the strategy that Government would adopt if: it decides to adopt a long term policy of moving toward the

introduction of a fully competitive electricity market; and Government decides to introduce the Single Wholesale Purchaser

structure after the expiry of the SCAs; and the Southern China market is not considered ready in 2008, but is

likely to be ready to be integrated with Hong Kong to form a competitive market at a later date.


introduction of a fully competitive electricity market; and Government considers that the Southern China market is ready to

be integrated with Hong Kong to form a competitive market based on the Multiple Wholesale Purchasers structure after 2008, and

Government decides to move directly from the existing structure to Multiple Wholesale Purchasers without first introducing the Single Wholesale Purchaser structure.


introduction of a fully competitive electricity market; and Government decides to introduce the Single Wholesale Purchaser

structure after the expiry of the SCAs; and the Southern China market is not considered ready in 2008, but is

likely to be ready to be integrated with Hong Kong to form a competitive market at a later date; and

Government decides to simultaneously introduce full wholesale (multiple purchasers) and retail competition once the Southern China electricity sector is considered sufficiently mature.

Based on the experience of other countries in introducing electricity sector competition (in a phased manner), such simultaneous introduction of full Wholesale and Retail Competition is considered inadvisable.


introduction of a fully competitive electricity market; and Government decides not to make any change at all until the

Southern China electricity sector is ready to be integrated into a competitive Hong Kong/Mainland China market at some point after 2008, and

Government decides to move directly from the existing structure to full Wholesale and Retail Competition without first introducing either the Single Wholesale Purchaser structure or the Multiple


115

Wholesale Purchasers structure. The Consultants consider such a trajectory both unnecessary and inadvisable.

Time Frames for Transition Through Market Models

3.4.63 The next step is to determine the potential time frame in which each type of market structure could be implemented, should Government decide that competition is the way to proceed.

3.4.64 Should Government decide to implement Market Structure 2, the benefits in terms of lower costs for new generation assets would be unlikely to be seen until after the completion of units 7 and 8 at Black Point. Due to the lead time sunk in land development, gas pipeline and transmission network at that site, alternative plant would most likely be unable to compete. The introduction of Market Structure 2 could therefore wait until the expiry of the current Scheme of Control agreement. The Consultants therefore suggest that 2009 would be a feasible date for the introduction of Market Structure 2, should Government decide on implementation.

3.4.65 The biggest single factor determining the timing of the implementation of a Multiple Wholesale Purchasers market would be the readiness of the Southern China electricity market to be integrated with that of Hong Kong on a competitive basis, thus providing additional, competitive sources of generation (see Section 4 for discussion of developments in Mainland China). The Consultants do not believe that this will occur within five years, but as Mainland China may develop rapidly, the Consultants recommend that the situation be reviewed in approximately three years to determine if the above recommendations should be updated. If the authorities in Mainland China are successful in meeting the general reform timetable that they have set for the electricity sector in Southern China, the integration of that market with Hong Kong to form a competitive wholesale electricity market may be possible.

3.4.66 Considering: an optimistic development and reform timetable for the electricity

sector in Southern China such that it matures to provide a source of generation competitors in ten years’ time, and

the fact that the current SCAs expire in 2008 the Consultants suggest 2009 as a provisional earliest possible start date for Multiple Wholesale Purchasers, should Government decide to move directly from Market Structure 1 to Market Structure 3. Alternatively, if the Southern China electricity sector is not sufficiently developed by 2009, or if Government does not wish to move directly from Market Structure 1 to Market Structure 3, Market Structure 2 could be implemented in 2009 and Market Structure 3 implemented at a later date, subject to the Hong Kong Government’s view on the readiness of the Southern China


116

electricity sector to be integrated with the Hong Kong electricity sector.

3.4.67 Market Structure 4 is a logical extension of Market Structure 3, the Consultants propose that the Market Structure 4 market structure could be implemented between ten and 20 years from now.

3.4.68 Making changes to the electricity sector is a relatively complex process, which requires a timeline involving: feasibility assessment (this Study); policy decisions; detailed planning; and implementation.

3.4.69 Table 3.4.b provides for an overview of the possible timing of the above market transition trajectories.

Table 3.4.bWindows of Opportunity for Implementation

Market Structure Window of Opportunity Opens 2 Single Wholesale Purchaser 2009: After expiry of current SCAs and

economic opportunities for independent tendering of new plant become available

3 Multiple Wholesale Purchasers 2009 or later, subject to the readiness of the Southern China electricity sector to provide generation competition

4 Full Wholesale and Retail Competition 2009 at the absolute earliest, subject to the readiness of the Southern China electricity sector to participate in generation competition with Hong Kong

3.4.70 The main message for Government policy-making from the analysis of possible competitive electricity market structures for Hong Kong and trajectories between them is that if Government develops a long term policy of implementing competition, then Market Structure 2 offers a common initial step in the most feasible of market transition trajectories.

3.4.71 Furthermore, introducing the Single Wholesale Purchaser market structure by independent tender of generating plant would not necessarily require major changes to the SCAs. It could be timed to coincide with the next major need for generating plant, or deferred until the expiry of the current SCAs.

3.5 Key Conclusions on Competition

3.5.1 The various alternative competitive market structures offer policy-makers a reasonable degree of flexibility. Given the existing electricity industry structure in Hong Kong and the experience with the introduction of competition around the world, it is clearly


117

advisable to implement electricity sector competition in stages, rather than moving to complete wholesale and retail competition all at once.

3.5.2 Government would need to monitor the implementation of competition, particularly in the early years and make any necessary adjustments to the market rules to ensure that the intended competitive pressures are functioning as intended. Those parts of the market that are not subject to genuine competitive pressures would need to remain under some form of regulation to protect the interests of consumers.

3.5.3 Experience elsewhere shows that wholesale competition is a pre-requisite for retail competition, but the reverse does not apply. It is therefore advisable to begin the implementation of competition at the wholesale end of the market, rather than the retail end. If a long-term policy of moving towards competition is chosen, then a number of paths from one market structure to another are possible. The Single Wholesale Purchaser market structure offers a common starting point in the most feasible of these alternative paths. This market structure could be implemented relatively smoothly after the expiry of the SCAs in 2009, assuming that the necessary detailed planning work is undertaken in the intervening years.


118

4 DEVELOPMENTS IN THE SOUTHERN CHINA ELECTRICITY SECTOR

4.1 Introduction

4.1.1 As discussed earlier in this report, the electricity supply industry in many parts of the world is moving towards separation of generation from transmission and distribution in a move towards competition and to gain consumer benefits. Mainland China is no exception and is implementing a philosophy called Separation of Government and Enterprises.

4.1.2 Specifically, the Central Government has determined that in order to introduce competition and improve benefits to consumers, the electricity sector must be restructured to provide a fair and open market system which is not directly run by the Government. Specifically, the system must: operate according to rules and regulations; have open and transparent pricing and access; separate government from involvement in the operation of the

electricity system; and maintain government as a regulator.

This Section

4.1.3 This section discusses: the present Guangdong and Southern China electricity supply

industry; restructuring plans for the Southern China electricity sector; the implications for Hong Kong, in the context of future electricity

sector competition.

4.2 Guangdong Electricity Supply Industry

Characteristics of the Sector

4.2.1 Electricity demand in Guangdong has been growing rapidly. In 1985-1997 the population in the province grew at 1.8% per annum and GDP at 15.3% per annum, while the generation capacity of the province grew at 18% per annum and electricity consumption at 14.7%. Table 4.2.a compares the Guangdong electricity supply industry with that of Hong Kong for 1997.


119

Table 4.2.a Comparison of Guangdong and Hong Kong Electricity Sectors (1997)

Indicator Unit Guangdong

HongKong

Total Generation Capacity MW 28 130 10 943

Maximum Load of Electricity Supply MW 16 530 7 271

Total Electricity Consumption GWh 92 000 33 000

Average Electricity Consumption/Person kWh/person

1 310 >4 000

Source: Guangdong Energy Technoeconomic Research Centre, HEC and CLP Annual Reports, Consultant’s calculations

Relevant Government Organisational Structure

4.2.2 The structure of the regulatory regime for electrical power is as complex as that of the industry itself. The foundations for regulatory reform in the power sector were provided by the Electricity Law which was passed in 1996. This law provides general provisions for the structure and regulation of China’s power sector. Please see Section 4.4 for a discussion of the law’s implications for the restructuring of the industry.

Generation Capacity in Surrounding Provinces

4.2.3 The three South-Western provinces (Guangxi, Yunnan and Guizhou) are projected to have large seasonal capacity surpluses which in the year 2000 could supply 1 660 MW, and 4 700 GWh, and in the year 2005, 2 440 MW and 7 300 GWh. Prices for wholesale purchase of these surpluses are also expected to be comparatively very low at 0.22 RMB (ie about HK$0.21) per kWh in daytime and 0.10 RMB (about HK$0.09) per kWh in the evening. These surpluses occur during the rainy season which lasts from April to September, the same time as Hong Kong’s peak demand.

4.2.4 In addition, further generating capacity from hydro power stations in the three South-Western provinces that have recently come on-line may allow transmission of power to Hong Kong more cost-effectively. Table 4.2.b shows the scale of excess capacity projected to be available during the rainy season.


120

Table 4.2.bCapacity in the Three Provinces of South-West China

1998 2000 2005

Province

SeasonUnit

wet dry wet dry wet dry

Yunnan MW 300 - 300 700 - 80 1 350 550GWh 1 000 - 800 2 200 - 200 4 000 1 000

Guizhou

MW 580 260 970 570 1320 630

GWh 1 500 0 800 2 600 1 500 4 100 1 700Guangxi

MW - 40 - 200 - 10 - 380 - 230 -1 020

GWh - 100 - 700 - 100 - 1 000

-800 - 2 500

Total MW 840 - 240 1 660 110 2 440 160GWh 2

400 - 700 4 700 300 7 300 2 000

Source: Guangdong Energy Technoeconomic Research Centre

Interconnection Between Guangdong Province and Hong Kong

4.2.5 To gain an understanding of the implications for interconnection with Guangdong Province, it is helpful to also understand how Guangdong is interconnected with other provinces.

Interconnection of Guangdong Grid with Neighbouring Areas

4.2.6 The Guangdong grid is an independent provincial level network, with a maximum transmission voltage of 500 kV. It forms a 500 kV circular circuit grid in the Pearl River Delta extending eastward to Shantou and westward to Moumin. The Guangdong grid is interconnected with the grids of Guangxi, Yunnan and Guizhou through the grid of the Southern Electricity Interconnection Company, which owns the interconnection cables of four provinces: Guangdong, Guangxi, Yunnan and Guizhou. Currently, the 500 kV First Tian-Guang Circuit is interconnected with the south-western grid. It has a length of 1 000 km and transmission capacity of 800 MW. The 500 kV Second Tian-Guang Circuit and DC 500 kV Tian-Guang Circuit will start to operate around the year 2000, so as to increase the transmission capacity from the West into Guangdong to more than 3 000 MW.

Electricity Exchanges between Guangdong and Hong Kong

4.2.7 Guangdong started to purchase electricity from Hong Kong in 1979, with the grids of Hong Kong and Guangdong connected by 66 kV transmission lines. As electricity purchases increased, the interconnection between the grids of Guangdong and Hong Kong


121

have been strengthened. Two 132 kV circuits connecting with Shekou and six 132 kV circuits connecting with Shenzhen have been built. In addition, four 400 kV circuits were built to transmit electricity to Hong Kong from Daya Bay in Shenzhen and a pumped storage power station in Guangzhou.

4.2.8 The quantity of electricity purchased from Hong Kong (CLP) varied between 160 and 320 GWh in 1979-83, rose to between 640 and 990 GWh in 1984 and 1985 and was maintained at 1 100 to 1 600 GWh in 1986-90. Purchases rose to 2 220 GWh in 1991, peaked at 5 400 GWh in 1993 and dropped to 1 260 and 1 230 GWh in 1994 and 1995, respectively. The Guangdong grid did not purchase electricity from Hong Kong in 1996 and 1997, as it became fully self-sufficient. As a point of reference, Hong Kong consumed over 31 000 GWh in 1996.

4.2.9 In 1985, Shekou in Shenzhen signed a ten-year electricity supply contract with CLP and the supply reached 400 GWh in 1993. In 1995, Shekou extended the contract another twenty years.

4.2.10 In 1994, CLP signed a 20 year contract obliging it to purchase 25% of the total electricity generated by Nuclear Group Company (25% is the share of the whole company held by CLP), and also to purchase 45% of total generation of Guangdong Nuclear Investment Company. Consequently, until the year 2014 CLP has to purchase 70% of the electricity generated by the 1 968 MW (gross capacity) Daya Bay Nuclear Plant, or approximately 7 000 to 8 000 GWh per annum. In 1994, CLP also acquired the right to use 50% of the 1 200 MW (gross capacity) Guangzhou Pumped Storage Station. The agreement will last until 2034. The quantity of electricity supplied to Hong Kong each year ranged from 200 to 500 GWh between 1994 and 1998.

4.3 Process of Restructuring

4.3.1 In 1998, the Central Government took its first step in implementing the concept of Separation of Government and Enterprises, or separation of the regulatory responsibilities and the responsibilities for the commercial operation of the power industry.

4.3.2 In early 1998, the Central Government dismantled the Ministry of Electric Power (MOEP) and allocated the regulatory responsibilities for the power sector to the State Economic and Trade Commission. The Central Government then established the State Power Corporation to administer state-wide transmission grids, operate cross-boundary transmission and operate state-owned entities.

4.3.3 There is a plan for provinces to follow the example of the Central Government to implement the concept of Separation of Government and Enterprises. However, the Central Government has not yet disclosed the provincial power sector restructuring plan or the timetable for opening up the electricity supply industry to competition.


122

Figure 4.3.a Planned Process of Restructuring for Guangdong Electricity Sector

Min

istry

of E

lectr

icPo

wer

Gua

ngdo

ng P

rovi

nce

Pow

er B

urea

uG

uang

dong

Pow

erH

oldi

ng C

ompa

ny

Regu

lato

ry

* Plan

ning

* Pro

ducti

on* P

ricin

g

Regu

lato

ry

* Plan

ning

* Pro

ducti

on* P

ricin

g

* Gen

erato

rs* T

rans

miss

ion G

rids

Com

mer

cial

Ope

ratio

nCo

mm

ercia

lO

pera

tion

* Gen

erato

rs* T

rans

miss

ion G

rids

Sepa

ratio

n of

gov

ernm

ent &

Ente

rpris

esSe

para

tion

of G

ener

ator

s &Tr

ansm

issio

n G

rids

1998

State

leve

l19

99 (e

xpec

ted)

Pro

vinc

e lev

el20

05 (e

xpec

ted)

R estru ctu rin g Fram ew ork


123

4.3.4 Guangdong Province, for example, is expected to undergo a restructuring of its power sector beginning in 1999. The Guangdong Province Power Bureau is expected to be dismantled and its regulatory responsibilities are expected to be allocated to the Guangdong Province Economic and Trade Commission. The Guangdong Power Holding Company (GPHC) is expected to take over the responsibilities of operating the province’s generators and transmission grids. GPHC is also expected to be monitored by the relevant government officials such as the Planning Commission, Economic Commission and Pricing Bureau.

4.3.5 During a transition phase from 1999 to 2005, GPHC is expected to legally become an enterprise with commercial responsibility for operating generators and transmission grids at both province and city levels. Guangdong Province Economic and Trade Commission will be responsible for planning and regulating the electricity sector in Guangdong.

4.3.6 Finally, one of the most crucial factors will most likely be the concept of Separation of Generators and Transmission Grids which will create two companies operating independently, one for generation and one for transmission. The concept is expected to be fully implemented by 2005 after the implementation of Separation of Government and Enterprises at province level.

4.3.7 Although not yet fully determined, Figure 4.3.a shows the general plan for the restructuring of the industry.

4.3.8 However, how the system will operate and what market model will be used is not yet clear. For example: will the system be operated on a Multiple Wholesale Purchasers,

Single Wholesale Purchaser or other basis? how will transmission prices be set? will there be a spot market? will generators be able to sign bi-lateral contracts?

4.3.9 Answers to these questions are critical to help the Hong Kong Government plan its own electricity supply industry development.

4.4 Implications for Hong Kong

4.4.1 The Consultants expect that, once the regulatory and other issues are resolved, some or part of the Southern China/Pearl River Delta region will become an integrated electricity grid similar to many other multi-jurisdiction grids around the world. Given the Basic Law and the ‘one country, two systems’ philosophy, it is also expected that administration and regulation of the electricity sector in Hong Kong and Mainland China will continue to be administered separately.


124

4.4.2 However, working on the assumption that closer co-operation between Hong Kong and Guangdong will ultimately lead to a more closely integrated electricity supply industry, there are a number of implications which need to be considered (all of which assume that reliability can be assured through a number of different mechanisms): supply from Mainland China providing additional generation

sources to spur competition; supply from Hong Kong to China (ie if Demand Side Management

and other initiatives take effect and Hong Kong continues to have excess capacity which can be sold to China);

potential for lower cost power supply; development of Hong Kong owned power stations, under SCA or

otherwise, in China; participation in a larger interconnected grid and the subsequent

reduction in reserve margins required; and a level playing field between Hong Kong and Mainland electric

utilities.

4.4.3 Each is discussed below.

Additional Generation Sources

4.4.4 One of the main barriers to the introduction of competition into the Hong Kong electricity supply sector appears to be the limited number of generating companies that currently exist in Hong Kong. However, as discussed above, there are numerous power stations in Guangdong province and many more are planned. In addition, the planned restructuring of the electricity supply industry there and entry by new companies may introduce even more separate entities which could act as competitors.

4.4.5 If the electricity sector in Guangdong is restructured such that the transmission system is operated independently and transparently, there is every reason to expect that Hong Kong’s electricity supply industry could take advantage of such a structure. Table 4.4.a details the implications of Mainland China’s restructured electricity sector for each of the competitive market models discussed in depth in Section 3.3.


125

Table 4.4.a Implications for Hong Kong of a Restructured Guangdong Electricity Sector

Hong Kong Market Model

Implications of Restructured Guangdong Electricity Sector

Vertically Integrated

HEC and CLP could purchase power through a least cost planning approach; and

Power could be sold on a long term contract basis.

Single Wholesale Purchaser

The Single Wholesale Purchaser could purchase power on a long term contract basis.

A restructured Guangdong electricity sector is not essential for the operation of a Single Wholesale Purchaser model in Hong Kong. However, a restructured market in China would be beneficial in providing a wider geographical area from which bids could be invited to supply power to the electricity companies in Hong Kong. It could also conceivably allow a variant of the Single Wholesale Purchaser model which replaces long-term power purchase agreements for power and energy with a combination of long term contracts for capacity and day-to-day bidding for the right to be dispatched.

Multiple Wholesale Purchasers/Retail Competition

Government could be assured that a large number of potential suppliers of power would help to ensure sufficient competition such that market domination (other than the issue of local Hong Kong Island market power) would not exist in generation;

suppliers in China could offer long term contracts to power marketers in Hong Kong; and

large customers, distributors, power marketers, etc could enter the spot market for economy and other transactions

Lower Cost Power Supply

4.4.6 Land, labour and many other inputs to the production and supply of electricity are less expensive in Guangdong than they are in Hong Kong. The wholesale price of power is also dependent on many other factors, including environment and safety requirements. However, the potential remains that the wholesale price of electricity may be lower in Guangdong than in Hong Kong. Plants in Guangdong may have higher forced outage rates than plants in Hong Kong, but to the extent that installed plant capacity was determined by system planning, this would be factored into the overall system reliability calculations, so sufficient reserve plant would be available to maintain system reliability at the desired level.

4.4.7 The Guangdong transmission system would also have access to low cost surplus hydropower electricity, estimated by some to be available for as little as 0.22 RMB per kWh (approximately HK$0.21 per kWh). As discussed above, this cheaper power could be tapped within any of the market models.


126

Development of Hong Kong Owned Power Stations in Guangdong

4.4.8 In addition to purchasing power from Guangdong utilities, there is also the possibility, given the extremely high costs of land in Hong Kong, that the two utilities could build power stations in Guangdong for the purpose of supplying electricity to Hong Kong.

4.4.9 CLP already owns assets regulated under its SCA within Guangdong and there is no reason to believe that further development in Guangdong of Hong Kong electricity utility owned assets could not occur. The cost-benefit trade-offs are relatively straightforward with the potential benefits of lower electricity costs weighed against the additional costs imposed upon the transmission system. In addition, transmission access issues must also be resolved.

Participation in a Larger Grid

4.4.10 A fundamental principle of this entire Study is that the larger the electricity grid within which a utility operates, the lower the overall reserve margins — spare capacity to cover emergencies — need to be to maintain a secure and reliable electricity system.

4.4.11 CLP and HEC already share spinning reserve requirements with GPHC and this is expected to continue. However, further reductions in reserve margins could be gained with greater integration into the Southern China power grid. For example, using a stricter criteria than Hong Kong currently uses (eg one day in ten year LOLP criteria), the reserve margin in the 50 000 MW Pennsylvania/New Jersey/Maryland power pool in the USA, for example, happens to be approximately 15%. This compares with reserve margins of 25%-30% in the HEC and CLP system, which is about 11 000 MW.

4.4.12 To fully understand the additional benefits to be gained in this area, further study will need to be performed. In addition, reliability issues will need to be carefully and specifically addressed.

Level Playing Field

4.4.13 Environmental, safety and other impacts of allowing power sales from Guangdong to Hong Kong have been raised, as many consider that the regulations in Guangdong and other provinces of China are not as stringent as in the SAR. Concerns have also been raised regarding the Hong Kong utilities’ rights of access — or ‘reciprocity’ — to the Southern China electricity market should Mainland firms have access to the Hong Kong market.


127

Environmental, Safety and Other Issues

4.4.14 Some argue that should a Guangdong power station be able to sell to Hong Kong, it could do so less expensively than Hong Kong utilities as it may not have to satisfy the same environmental, safety and other requirements that Hong Kong power stations must meet. If the Hong Kong Government or utilities want to impose environmental restrictions on power sold to Hong Kong, there are a number of possible methods that could be adopted: only allow sales from power stations that meet specific

environmental, safety and other requirements; require participation in the project from the Hong Kong utilities

and require them to ensure that environmental, safety and other performance is maintained; or

only allow new power stations to sell to Hong Kong.

4.4.15 Introduction of competitive supply from China to help introduce competition into the Hong Kong electricity sector would require the balance between gaining economic benefits for consumers and minimising environmental and health and safety concerns to be clearly thought through. Furthermore: some benefits of a competitive power market may be lost by

excluding certain participants, especially those which have low prices;

responsibility for monitoring the power station to ensure that it is meeting requirements must be specified; and

penalties and enforcement mechanisms must be also specified.

4.4.16 The Mainland China electricity sector is evolving rapidly and the Consultants do not foresee the need to make determinations on the above issues until a time closer to the point that further supplies to Hong Kong are considered likely.

Reciprocity

4.4.17 It might be argued that, should Mainland firms have access to the Hong Kong electricity market, HEC and CLP should, in the name of fairness, have access to the Southern China market. There are advantages and disadvantages to this line of reasoning. From an economic perspective, if full cross-boundary competition were to be allowed, it would be preferable if the entire market were completely transparent with all generators selling within the same market rules. This is generally the objective of other countries’ efforts at electricity sector reform and the introduction of competition.

4.4.18 However, there may be economic disadvantages for Hong Kong consumers should the lack of reciprocity be used as a reason to


128

prohibit introduction of additional competitors from Mainland China. For example, if the Mainland generators could produce power in a method acceptable to Hong Kong Government (ie meeting environmental, safety and other criteria) at a lower price than HEC or CLP, some groups might ask why Hong Kong consumers should be forced to pay higher electricity prices? Although the economic theory of trade would support consumer groups, it needs to be recognised that there are broader considerations at work, such as equity in trading relationships. Therefore, decisions about reciprocity need to be considered carefully on a case-by-case basis.

4.4.19 In the event of cross-boundary power sales, it would be preferable if there is complete reciprocity of power sales between Hong Kong and Mainland China. However, should there not be complete reciprocity, Government may have to make decisions that may ultimately favour either consumers, the utilities or broader interest groups over the others, taking into consideration such issues as who receives the benefits, job losses or creation, and other issues that would need further analysis.

4.4.20 Although the Consultants consider that effective competition will not be practical in Hong Kong until there is greater integration with the Mainland electricity market, there are a number of significant issues that must be resolved prior to full integration. Hence the Consultants recommend that the Southern China electricity sector be monitored by Government and a review be undertaken in three to five years time to re-assess these issues and their implications for a competitive power market in Hong Kong.

4.5 Key Conclusions on Co-ordination with Southern China

4.5.1 The eventual inclusion in a future competitive Hong Kong electricity market of electricity generation companies in Southern China is considered necessary for the establishment of genuine wholesale competition, due to the practical limitations for sufficient competing entities to be located within Hong Kong. Such inclusion is likely to be possible in the future. It would first require a number of commercial, legal, technical and environmental issues to be agreed between the relevant parties.

4.5.2 Although Hong Kong is part of China, it has a separate legal and commercial system and its electricity system is at a more advanced stage of development. Nevertheless, the appropriate co-operative relationships between Hong Kong and Southern China should be established in the short term. A broad review (but with more detail than was possible in this feasibility Study) of the technical, commercial legal and regulatory requirements for a Hong Kong-Southern China competitive wholesale electricity market should be undertaken. The general situation should be monitored and reviewed periodically.


129


130

5 CONCLUSIONS AND RECOMMENDATIONS

5.1 Introduction

5.1.1 The conclusions and recommendations presented in this Section respond to the Terms of Reference. However — as pointed out in the Section 1 Introduction and Overview of this report — there are a number of topics intimately inter-related with the two main Study topics of interconnection and competition. The conclusions and recommendations presented here include consideration of these issues to give the reader a wider appreciation of the issues that make up the overall context.

5.1.2 It is clear from the stated objective of the Terms of Reference,(1) that a “big picture” view of the electricity supply industry in Hong Kong is required: The Consultants will be required to provide independent findings and recommendations sufficient to enable the Government to formulate a substantive response to public concern on these issues.

5.1.3 Within that context, the Consultants believe that the focus should be on implementing the industry structure most likely to deliver benefits in the medium- and long-term. The optimal industry structure could be developed by implementation of competition in a series of phases. This cannot be done immediately for a number of reasons explained in the report and summarised below, and it will require thought and careful design and implementation, because there is no international model of an electricity sector where competition has been introduced that is exactly like Hong Kong.

5.1.4 The interconnection analysis provides a feasibility level quantification of benefits. The results need to be then considered within the context of the current industry situation and the prevailing legal framework. Large and sudden changes based only on a narrow technical and economic analysis would have much wider ramifications which may alter the industry structure significantly in ways not directly assessable within this Study and which may reduce other options for restructuring in the future. These would need to be further considered within the wider Government policy context, deliberated and the public consulted.

1(?) Included in Section 1 — Introduction and Overview.


131

This Section

5.1.5 This Section is structured as follows: conclusions on the technical feasibility of making increased use

of interconnection between HEC and CLP; conclusions from the results of the economic analysis of

interconnection/generation scenarios, including a discussion of the most important factors that determine the result;

conclusions from the results of the financial and tariff analysis of interconnection/generation scenarios;

conclusions on logistical considerations associated with implementing increased interconnection;

conclusions from the results of the assessment of options for the introduction of competition;

conclusions on regulatory requirements associated with each of the scenarios;

conclusions from the results of the assessment of the Southern China electricity sector;

conclusions on the wider implications of the interconnection/generation scenarios, particularly with a view to the possible future introduction of competition; and

recommendations.

5.2 Conclusions on the Technical Feasibility of Increased Interconnection

5.2.1 The flow of power on the three existing interconnector circuits is not evenly distributed (it is “unbalanced”) and this reduces the transfer capability of the existing interconnector. It is thought that the cost of balancing the circuits would be quite small and would have the effect of increasing the flow on the three existing interconnector circuits to approximately 430 MW.

5.2.2 Increased interconnection capacity is technically feasible, although there may be some logistical and timing constraints which need to be studied further, particularly due to the congestion in the harbour, the landing points at the Central-WanChai Reclamation and cable routes on the Hong Kong side. These are discussed in more detail after the presentation of the economic results.

5.2.3 The existing system is at present marginally stable, with just enough damping, but under certain conditions may be prone to low frequency oscillations. Each additional generation unit would slightly decrease system damping, which with the existing interconnection capacity would tend to increase the susceptibility of the system to inter-area low frequency oscillation problems. As generation units are added to the system therefore, increased interconnection


132

capacity will eventually be needed to increase the damping,(1) although this would not necessarily be required before the next individual unit addition.

5.2.4 It is further expected that increasing the interconnection capacity would also increase the overall generation and transmission reliability of the system, by introducing an additional strong supply path to Hong Kong Island in addition to the existing paths from the Lamma power station and from Kowloon via the existing relatively small interconnector. However, a full Level I (generation) and Level II (transmission) reliability analysis of the interconnected HEC-CLP system would be required to completely confirm this.

5.3 Conclusions on Economic Analysis of Interconnection/Generation Scenarios

5.3.1 The analysis of interconnection/generation scenarios assumed that balancing of the existing interconnector would be undertaken immediately and the highest (conservative) cost option was assigned to this in the economic analysis.

5.3.2 The quantitative feasibility-level economic analysis in this Study focused on overall system-wide economic benefits from reductions in generation capacity needs that would be potentially available under Joint Planning — which would require increased interconnection used for Firm Capacity support between areas. Quantifying precisely any additional benefits potentially available under least cost Joint Dispatch of generation plant — which would require use of increased interconnection for Firm Power transfer — would be a more complex task requiring the adoption of a large number of assumptions. Inclusion of these additional benefits would be expected to increase the potential economic benefits somewhat. However, the qualitative analysis conducted indicated that the these potential benefits from Joint Dispatch would most likely be significantly smaller in magnitude than the quantified economic benefits potentially available from Joint Planning. Inclusion of potential Joint Dispatch economic benefits would not therefore be expected to substantially alter the results of the analysis, and would not alter the conclusions.(2)

5.3.3 The results show that overall economic benefits relative to the base case would potentially be available from an

1(?) The increase in system damping that would be provided by the second large interconnector in interconnection Scenario B would help overcome the potential with the existing interconnection for low frequency oscillation problems.2(?) The discussion of Joint Dispatch here does not include the potential effect of competitive forces on generation costs. Although implementing competition in electricity generation requires the joint dispatch of plant for the operation of that competitive market — which would be expected to drive down generation-level costs and prices — this does not apply in reverse. That is, Joint Dispatch — between the two Hong Kong utilities in the limited, economic or “least cost” sense employed here — does not necessarily entail the operation of competitive forces. Competition was considered separately, and the conclusions on that topic are presented below in sub-section 5.7.


133

interconnection/generation scenario with a significantly upgraded HEC-CLP interconnection. These overall economic benefits would have a present value of $347 million, $562 million and $896 million, in the periods up to 2008, 2018 and 2028, respectively, using a real discount rate of 4% per annum. These savings would be equivalent to approximately five per cent of the relevant incremental present value of capital expenditure on generation that would be required without any upgrade of the existing interconnection.

5.3.4 The savings described would be available from interconnection upgrade scenario B, whereby: the interconnection is upgraded via the addition of a high voltage

(275 kV) large capacity (2 x 550 MVA) second interconnector by the year 2004; and

additional generation would be installed in 2006 and 2007 in the form of Black Point units 7 and 8 (two 312.5 MW combined cycle gas turbines).(1)

5.3.5 The analysis also shows that a small amount of additional generation capacity would be needed in 2008 and 2009(2) and that the most economical way of providing this would be several small gas turbines on an available site. No other additional generation capacity would be required in the period up to 2008.

5.3.6 This result is entirely consistent with the result of the technical analysis that interconnection case B represents the best option.

5.3.7 Savings from interconnection upgrade scenario A, would be small. In interconnection upgrade scenario A: the existing lower voltage (132 kV) interconnector would be

upgraded via the addition of a fourth 240 MVA tie by the year 2004; and

additional generation would be installed with Black Point unit 7 and the first Lamma extension unit (Lamma 9 a nominal 300 MW combined cycle gas turbine) in 2006, Black Point unit 8 in 2008 and the second Lamma extension unit in 2008.

1(?) CLP has contracts to purchase these two generation units, which were originally scheduled for installation in 2000 and 2001. All of the civil works and transmission connections for these units are already in place. Based on agreement with Government, the two units have been deferred by three years each to 2003 and 2004, with the possibility of further deferral to 2005 and 2006. The timing of the installation of these units in each interconnection case in the Study analysis was determined by the planning criteria and economic considerations from an overall system point of view, not by contractual considerations. However, it should be noted that CLP’s agreement with Government under the Scheme of Control Agreement — on which they based their contracts with suppliers — is legally binding and any attempt to alter these agreements would have to be agreed to by the suppliers and would have compensation implications. 2(?) However, as noted in the introduction, the Study analysis does not include the effect of the CLP-GPHC interconnection in the calculation of generation adequacy for assessment of Level I (generation) reliability. If this was considered, it is possible that the need for this additional generation may be delayed somewhat.


134

5.3.8 Analysis shows that it would not be economic to upgrade the existing 132 kV interconnector once the second, higher capacity interconnector had been upgraded.

5.3.9 Therefore interconnection upgrade scenario B is the most economic alternative as well as the most sound option from a technical point of view. There are several factors that lead to this economic result. In general, interconnection allows generation reliability levels to

be met with less installed capacity. The next two generation additions at CLP are available at very

low incremental cost, because all of the investment at the Black Point site — apart from the installation of the actual combined cycle gas turbine units themselves — has already been made. The analysis is conservative in valuing the benefits of interconnection because it counts the cost of the units themselves, although an economic case could be made that, since CLP are contractually bound to purchase the units, with Government agreement, there is no incremental cost for the units.

Sufficient land is also available at the existing Black Point platform to accommodate three small open-cycle gas turbines and one large combined-cycle unit with a total capacity of some 1 260 MW. Sufficient transmission capacity is also available to deliver the output from these units, after allowance for a severe contingency where the two largest circuits are forced out of service such as by collapse of the supporting tower.

The next generation additions at HEC will require land reclamation, construction of a submarine gas pipe and laying submarine transmission cables. The cost estimate for these is in the order of $3.4 billion, before counting the cost of the generating units themselves. The analysis of potential economic benefits is technically and economically conservative in assuming that the transmission cables — accounting for $1.7 billion or half of the $3.4 billion — cannot be deferred beyond the base case installation date of 2004. This assumption ensures that there would be no local transmission constraint on Lamma generation, so that the HEC-CLP interconnection would always be the transmission constraint between HEC generation and CLP load as modelled in the LOLP analysis. The assumption also means that the calculated benefits for the period up to 2008 are not contingent on establishing use of the interconnection for Firm Power transfer. If subsequent, more detailed system analysis indicates that it is possible to defer this transmission investment and it is also possible to establish Firm Power arrangements over the HEC-CLP interconnection before 2008, then the potential economic benefits of interconnection/generation Scenario B would increase by approximately $350 million.

Additions at CLP after the three small gas turbines and one large combined-cycle unit mentioned above would require additional


135

cut-and-fill site formation and additional transmission capacity via overhead lines.

Overall economic analysis compares incremental costs. Costs already incurred are considered “sunk” in economic terms. The equivalent of the high initial costs faced by HEC — site formation, construction of a submarine gas pipeline(1) and connection with the transmission network — have already been sunk by CLP.(2)

In addition to consideration of the stage of investment described above whereby some substantial site and transmission costs at CLP are already sunk, CLP currently has some cost advantages over HEC. Land formation of an artificial island at the proposed site within HEC would tend to be somewhat more expensive than the cut-and-fill technique possible at the site available to CLP. Power from the CLP site is delivered via overhead lines, whereas the HEC site requires submarine cables.

Some of the benefits calculated for increased interconnection come from the general deferral of plant, but most come from the opportunity to defer construction of high initial cost new generation at HEC that would be made possible with installation of increased interconnection capacity along with the installation of Black Point units 7 and 8 and substantial further additional capacity at that site at low incremental cost.

5.3.10 Under the interconnection/generation scenarios considered above, HEC would rely on CLP primarily for reserve generation support, rather than for plant dispatch. However, in the years around 2008, when the HEC’s own reserve plant levels would be low, HEC would need to use the interconnection for Firm Power transfer from CLP at least during peak periods, depending on the flexibility with which HEC could schedule maintenance of its units. The detailed operational implications of such a scenario would need to be reviewed with consideration of the concerns that either utility may have about their supply reliability being dependent to a significant extent on generation support from another company. These issues would require, among other things, the approach to generation capacity support obligations between companies to be clearly defined at a detailed operational level, and would most likely require substantial changes to the existing Scheme of Control and Interconnection Agreements. All of these would need to be carefully considered in any decisions taken.

1(?) In CLP’s case, the capital cost of all but the last 5 km of the gas pipeline is expressed in the price and contract terms with the gas supplier. 2(?) It is estimated that in the hypothetical case where the existing site and transmission assets (that are able to accommodate several additional generation units beyond Black Point unit C8) were not present, the present value of the potential benefits of increased interconnection would be up to $400 million less, with the actual value dependent on the assumptions made about the costs and timing of the investments. The analysis assumes that there is sufficient capacity available in the existing pipeline and that there will be sufficient gas available from the existing source to supply several CLP generation units in addition to the eight units in the first phase of the Black Point power station. If this turned out not to be the case, then it is estimated that bringing forward investment in an alternative gas supply would reduce the present value of the potential benefits of increased interconnection by about $15 million.


136

5.3.11 Additional economic benefits from use of the interconnection for Firm Power transfer would be available from joint dispatch of HEC and CLP plant in the most economic order. However, implementing Firm Power arrangements in the period before 2008 to obtain some (probably small) additional benefits from Joint Dispatch in the short term would require more significant changes to the current arrangements, particularly to the Interconnector Agreement, than would be necessary if implementing Joint Planning only in that period.

5.3.12 Furthermore, the dispatch of plant would be driven in part by contractual considerations, including: must-run operational constraint of the Daya Bay nuclear plant and

CLP’s commitment to take a significant amount of energy from that plant; and

must run status of plants using gas (Black Point combined cycle units) under take-or-pay contracts.

These contracts have been struck under the framework of the Scheme of Control Agreement with Government. The nuclear contract is due to expire in 2014 and the gas contract at the end of 2015.

5.3.13 Additional determinants on dispatch may be applied on environmental grounds, such that coal plant with flue gas desulphurisation (FGD) must be dispatched before non-FGD coal plant. This would be the result of environmental policy to avoid non-FGD plant displacing FGD plant based on dispatch cost. Similarly, environmental concerns would dictate that coal-fired plant should not simply be able to displace gas-fired plant based on dispatch cost, but given the fact that existing take-or-pay contracts would determine the dispatch of gas plant in the near term, this consideration would be redundant.

5.3.14 Use of the interconnector for Firm Power transfer would require a stable system free from low frequency oscillation problems. There is some risk of low frequency oscillation problems with the existing interconnection but, as concluded above, increased system damping would be a by-product of the upgrade of the interconnection, thereby resolving this area of concern. The implications of this for the economic analysis are that Firm Power arrangements are not considered technically feasible before 2004, as this is the earliest possible date for upgrade of the interconnection.

5.3.15 The existing contractual constraints on full economic joint dispatch will persist into the medium-term, when the CLP nuclear and gas contracts are due to expire. Generation companies may choose or find it necessary to enter into long-term power purchase contracts in future. However, if economic dispatch is eventually determined by the operation of a market with fully competing wholesale suppliers, then those generation companies’ revenues would be determined by the prevailing wholesale market prices, rather than determined by


137

the process embodied in the Scheme of Control Agreements as they are for the vertically-integrated companies at present. The generation companies would therefore have to take into account the commercial risks associated with their decision-making, because in such an environment these risks could not be automatically passed on to the end consumers. Managing such risks is part of the normal business environment for companies in other industries. This is also the case for electricity generation companies in places where they are subject to full competition. In some countries, electricity sector commitments under existing fuel contracts are phased out gradually, and the companies increasingly subjected to market rules over time, rather than waiting for the contract periods to expire. In any case, it would be advisable for Government to provide the electricity sector with a suitable period of notice about the future market structure planned, so that the industry can take this into account when entering into fuel supply contracts.

5.4 Conclusions on Logistical Considerations Associated with Implementing Increased Interconnection

5.4.1 The specific calculation in this Study of overall economic benefits available from increasing and making increased use of HEC-CLP interconnection are predicated on the assumption that the specified interconnection upgrades could be installed at the specified dates. If interconnection capacity was increased at a later time, the potential economic benefits for that scenario would need to be analysed. It is expected that such a scenario would offer potential economic benefits, but that they would be significantly less than those estimated for Scenario B in this Study.

5.4.2 Most of the required initial pre-construction steps for constructing additional generation at HEC have already been completed. These include: selection of a specific site, selection of a submarine cable route, selection of a submarine gas pipeline route and the associated environmental impact assessments. This took approximately four years to achieve.

5.4.3 None of the initial pre-construction steps for increasing the HEC-CLP interconnection capacity have been conducted. These include selection of a specific submarine cable route, identification of suitable landing points at the Hong Kong Island and Kowloon Peninsula ends, identification of cable routes for the land portion, identification of suitable substation facilities at each end and application and Government approval for each route and site. It is possible that this process could take two to three years, given that building an interconnector across the congested harbour will pose challenges and there are constraints on cable route and landing points. The actual time taken would depend mainly on resolving detailed technical and implementation issues related, amongst other things, to land-side route selection — particularly in the congested urban areas — and in satisfying the various Statutory and


138

Administrative approval processes.(1) It should be possible, once the planning process has been completed and the necessary approvals obtained, to complete construction of the actual interconnection within two to three years.

5.4.4 There are risks associated with each of these resource options. The interconnection requires that suitable route and land access be obtained in the required time. It would also require both companies to agree to share the costs, build the interconnector and make the connection with their respective systems. The generation option requires that a new LNG terminal in Southern China be constructed to coincide with the timing of the new generation units. The plan adopted would need to take into account contingencies in each schedule, in order to maximise flexibility and minimise the risk of capacity being unavailable when it is needed.

5.5 Wider Implications of the Interconnection/Generation Scenarios

5.5.1 There are a number of wider considerations — beyond the direct economic analysis of potential benefits — with respect to increased interconnection. A number of these considerations act in opposing directions, such that they must be carefully balanced so that the appropriate solution can be found. There is no simple, straight-forward and easy answer to the question.

5.5.2 In addition to economic and consumer benefits, these considerations include: the effect on the utilities’ commercial positions — which are

planned on the basis of the existing Scheme of Control Agreements — of making what amounts to a relatively significant change in the planning approach from one based on two quite independent systems with only a low level of interconnection to one based on quite closely integrated joint planning with a much higher level of interconnection;

the relative competitive positions of the two utilities at the planned date for introduction of competition.

5.5.3 Under the current Scheme of Control Agreements, which run to 2008, the Government cannot require the utilities to take steps such as construction of increased interconnection capacity and any modifications to their Interconnection Agreement. Implementation of such changes before 2008 would require voluntary acceptance by the utilities. Post-2008 a new mechanism would be required to allow the Government, if it so desired in the interests of consumer benefits, to require the utilities to adopt system measures such as increased interconnection.

1(?) HEC have cited the addition of Sheung Wan - Kennedy Road circuits as an example of long approval times, for which it apparently “took some 6 years to get approval from some 20 government departments.”


139

5.5.4 Making increased use of interconnection to defer construction of additional generation at HEC could deliver economic benefits to Hong Kong, which would translate to consumer benefits under the present SCA structure, and such reserve capacity-related benefits from Firm Capacity with increased interconnection can be considered to fit within the scope of the existing Interconnection Agreement between the utilities. However, the statement in the existing Interconnection Agreement, that “Neither party shall install, cause to be installed, acquire, or cause to be acquired excess generating capacity beyond its own need in the attempt to supply the other party's load” may be interpreted to exclude use of the interconnection for Firm Power transfer, particularly where such arrangements lead to greater energy exchange in one direction than in the other direction. It should be noted that some Firm Power transfer of power across the interconnection would be necessary in some years at least during peak hours. This would be the case in, for example, interconnection/generation Scenario B, under which HEC installed capacity would be less than HEC peak load in some future years, such that CLP plant would be needed to meet HEC peak load. The need to rely to this extent on generation support from a separate company to meet reliability requirements has been identified by HEC as a matter of serious concern, particularly in the context of the existing Scheme of Control Agreements, under which each company is responsible for providing reliable supply to its customers. In the same scenario, the low levels of the shared reserve generation capacity located within HEC would also tend to result in more frequent transfers of power across the interconnection under its main role in providing Firm Capacity generation support than would be the case in other scenarios where a greater proportion of shared reserve plant was located within HEC.

5.5.5 The results of the economic analysis are to a significant extent a reflection of the relative stages of generation development cycles between the two utilities and to a certain extent a reflection of the high generation plant siting costs within HEC. They reflect the fact that a comparison is being made between: the last two steps of one large stage of generation investment

stage (8 x 312.5 MW of combined cycle gas turbine units at Black Point) and the addition of four more generation units totalling 1 260 MW within already available site space and transmission capacity; and

the commencement of another large generation investment stage (6 x ~300 MW combined cycle gas turbine units at the planned Lamma extension), which requires substantial up-front investment to form a site.

5.5.6 Making use of low incremental generation expansion options at CLP made possible by existing commitments and additional available existing land and transmission capacity to achieve a system-wide least-cost outcome could be perceived from a commercial perspective as unduly favouring CLP at the expense of HEC.


140

5.5.7 HEC has expressed the concern that installation of a larger interconnector and making increased use of HEC-CLP interconnection to allow the installation of Black Point units 7 and 8 and subsequent units to capture system-wide economic benefits would unfairly disadvantage them from a commercial point of view under the structure embodied in the existing Scheme of Control Agreements, whereby utility profits are proportionately related to the value of fixed assets. It may also be seen as weakening their position in a future competitive market.

5.5.8 CLP has expressed the concern that they have a legally binding commitment to purchase the final two units for the Black Point power station based on previous Government approvals, and that there is no provision in the existing contract for deferring these units further.

5.5.9 It is only to be expected that both electricity companies would defend their commercial interests and what they see as their business rights under their existing Scheme of Control Agreements with Government. The Consultants can appreciate the merits of both points of view, as well as the opportunities available for economic and consumer benefits. The challenge arising from this Study for the two utilities and for Government is to find the most appropriate resolution of these conflicting requirements.

5.5.10 The resolution eventually settled upon in the near term must be compatible with Hong Kong’s long term strategy for development of the electricity sector. If Hong Kong decides that the long-term strategy is to include moving towards the introduction of competition after the expiry of the existing Scheme of Control Agreements in 2008, then a number of considerations need to be weighed carefully.

5.5.11 Like all prices, tariffs are comprised of costs and margins. Competitive pressures tend to exert downward pressure on both costs and margins. There may be a risk that implementing increased interconnection between HEC and CLP, deferring construction of additional generation and implementing joint plant dispatch in the near term — which would deliver economic benefits by reducing costs hence tariffs under the SCA — may also have the effect of strengthening CLP and weakening HEC from the point of view of future introduction of competition. This may allow CLP to dominate the market, therefore hindering the effective operation of the competitive market pressures that are required to limit margins and thus work against longer-term consumer interests.

5.5.12 However, the Consultants conclude that increased interconnection capacity will eventually be necessary for competition and technical reasons. With respect to competitive market considerations, the existing small interconnection capacity (which will be even smaller in relative terms) may represent a constraint on competition for supply of electricity to customers on Hong Kong Island in a future


141

competitive market, thereby allowing HEC to exercise market dominance in that geographical area. With respect to technical considerations, increased interconnection capacity will eventually be needed as generation units are added to ensure sufficient damping to provide an interconnection that is not susceptible to inter-area low frequency oscillations so that it could be used for reliable power transfer. Such an interconnection would be necessary should a competitive market be implemented in the future. Therefore, if its capacity was not upgraded prior to the introduction of competition, the existing HEC-CLP interconnection would represent a technical barrier to the implementation of competition, as customers on Hong Kong Island and generation plant at Lamma Island could not participate in a competitive market.

5.6 Conclusions on Financial and Tariff Analysis of Interconnection Cases

5.6.1 The reduced revenue requirements — which would flow through into consumer bill savings via tariff reductions — have been calculated under the existing Scheme of Control Agreements using the economic benefits calculated for each of the generation and interconnection expansion scenarios, assuming: CLP finances new assets with 20% equity and 80% debt, with the

principal repaid in equal instalments over ten years; HEC finances new assets with 100% debt, with the principal

repaid in equal instalments over thirty years; assets are depreciated on a straight line basis over thirty years; an interest rate on borrowed capital of 8% per annum; a real discount rate of 4% per annum; and an implied annual inflation rate for prices of fixed assets of 3.85%

per annum.(1)

5.6.2 The net present value of consolidated reduction in annual revenue requirements under the existing Scheme of Control framework of scenario B compared with the base case would be $1 236 million in the period up to 2008. This value is calculated from the reduction in the annual revenue requirements for Scenario B relative to the base case under the framework of the existing Scheme of Control Agreements. If the existing Scheme of Control accounting framework is similarly applied to the economic results for the period up to 2018 and up to 2028, the corresponding consolidated reductions in revenue requirements are $994 million and $924 million, respectively. For illustrative purposes, the simple average of the consolidated annual tariff reductions associated with the consolidated reductions in revenue requirements for the period up to 2008 is 0.40 ¢/kWh. The Development Fund provided for

1(?) A nominal discount rate of 8% p.a. and a real discount rate of 4% p.a. as used in the economic analysis, implies an inflation rate of 3.85% p.a.


142

under the Scheme of Control could be used to smooth the tariff adjustment effects of the effect of an interconnection/generation scenario on annual utility revenue requirements.

5.6.3 However, these reductions in revenue requirements would not be evenly distributed between HEC and CLP. If, for example, Scenario B was implemented under the existing Scheme of Control Agreements without any change or adjustment to those agreements, then in the period to 2008 HEC would experience a decrease in annual revenue requirements relative to the base case with a present value of $2 003 million, while CLP would experience an increase in annual revenue requirements relative to the base case with a present value of $767 million. This would correspond to an average tariff decrease for HEC customers in the period to 2008 of 2.21 ¢/kWh and an average tariff increase for CLP customers of 0.35 ¢/kWh.

5.6.4 The system-wide economic savings and the consolidated customer bill savings result from the lower overall cost of interconnection/generation investment scenario B compared with the base case investment scenario. However, scenario B would result in investment in fixed assets within CLP in the period up to 2008 with a slightly higher net present value than the base case (hence the upward effect on CLP tariffs under the SCA). But investment in fixed assets within HEC would be substantially lower in scenario B than in the base case investment scenario (hence the downward effect on HEC tariffs under the SCA). The reason that the tariff decrease for HEC customers under the existing Scheme of Control Agreement would be large and the corresponding tariff increase for CLP customers would be small is that energy sales in HEC are much smaller than in CLP.

5.6.5 These results highlight precisely the important point(1) that it is difficult to find a fit between the existing regulatory arrangements — which were originally developed in the context of two separate, independent utilities — and least-cost resource options available from co-ordination and joint planning between the two utility systems.

5.7 Conclusions on Analysis of Future Competition Options

5.7.1 The introduction of competition would offer a way to expand the range of electricity supply options for Hong Kong in the medium- to long-term, provide an environment where Government could eventually be relieved of the task of directly regulating utility profits for some utility industry activities, new incentives would emerge for the utilities to improve their overall performance and efficiency, while at the same time serving customers’ electricity needs at lower overall cost.

1(?) As suggested under Implications of the Current Situation for this Study in Section 1 — Introduction and Overview.


143

Assessment of the Southern China Electricity Sector

5.7.2 Development of the South China market has important implications and has the potential to offer significant future opportunities for Hong Kong. However, these opportunities are dependent on the successful restructuring of the electricity supply industry in Southern China. Without successful restructuring, there would exist no practical opportunity to implement a Multiple Wholesale Purchaser model in Hong Kong.

5.7.3 The Mainland China authorities have announced plans for the reform of the electricity supply industry in Southern China, but it is not yet clear at this stage precisely how and when these plans will develop into reality. Electricity sector reform in a number of advanced industrial economies has tended to take considerably longer than the planners originally envisaged (Australia, New Zealand, and the United States are all examples of this). Therefore, it is highly likely that the reform plans in China, which has an electricity sector that is still developing and faces many technical challenges, will take considerably longer than the authorities’ plans might suggest.

5.7.4 It is clear from the experience internationally that any move towards the introduction of electricity sector competition must take account of the existing industry ownership, organisational and regulatory structures in the particular country or countries in question and the process of change must be tailored for each particular place. However, the international review of electricity sector competition has shown that there are examples of places with small markets and limited potential for generation competition — such as Northern Ireland and Scotland — where a carefully designed market can be introduced to allow competitive pressures in electricity generation to replace regulation.

Feasible Market Structures, Transition Paths and Timing

5.7.5 No major reforms of the existing market structure are advisable until after 2008, since there is no real potential to introduce generation level competition before that time. This situation also happens to be consistent with the fact that the Scheme of Control Agreements do not expire until 2008 and terminating them prematurely would raise questions of utility compensation for investments made with Government approval under that framework.

5.7.6 After 2008, Market Structure 2 — Single Wholesale Purchaser would be feasible, with new generation independently tendered. Given the lead time, including site search and selection and the time frame for construction involved in investing in new generation, the tender process would need to be initiated well before the date of the need for new generation. If competitively tendered new generation was required in 2009 for example, the tenders would need to be called around 2003 to 2004. Some form of regulatory control would continue to be required under this market structure, but the form of


144

this regulation would need to be different from the Scheme of Control Agreement in a number of respects.

5.7.7 Once the electricity sector in Southern China is sufficiently developed, it would become possible to introduce Market Structure 3 — Multiple Wholesale Purchasers, whereby generators in Southern China and Hong Kong compete with one another.

5.7.8 Once it is possible to introduce full wholesale competition under Market Structure 3, it would also become possible to introduce Market Structure 4 —Wholesale and Retail Competition.

5.7.9 There are a number of possible transition paths from the current market structure to the alternatives. However, the Consultants consider that not all of these trajectories are advisable. Based on the experience of other countries in introducing electricity sector competition (in a phased manner), the simultaneous introduction of full Wholesale and Retail Competition — moving directly from a Single Wholesale Purchaser structure — is considered inadvisable. Similarly, a market transition trajectory that proceeds directly from the existing Vertically Integrated structure to full Wholesale and Retail Competition is both unnecessary and inadvisable.

5.7.10 Therefore, only the most promising three trajectories for market transition are considered here. These are: Trajectory 1 — from Market Structure 1 to 2 — which would be

chosen by Government if it decides on limited reform, but decides not to move to the introduction of full competitive markets.

Trajectory 2 — from Market Structure 1 through 2 and 3 to 4 — which represents the strategy that Government would adopt if it decides to adopt a long term policy of moving toward the introduction of a fully competitive electricity market; and decides to introduce the Single Wholesale Purchaser structure after the expiry of the SCAs; and the Southern China market is not considered ready in 2008, but is ready to be integrated with Hong Kong to form a competitive market at a later date.

Trajectory 3 — from Market Structure 1 direct to 3 and then to 4 — represents the strategy that Government would adopt if it decides to adopt a long term policy of moving toward the introduction of a fully competitive electricity market; and Government considers that the Southern China market is ready to be integrated with Hong Kong to form a competitive market based on the Multiple Wholesale Purchasers structure after 2008, and Government decides to move directly from the existing structure to Multiple Wholesale Purchasers without first introducing the Single Wholesale Purchaser structure.

5.7.11 It is clear from these options that there exists a reasonable degree of flexibility for policy-makers. The main message for Government policy-making from the analysis of possible competitive electricity market structures for Hong Kong and trajectories between them is


145

that if Government develops a long term policy of implementing competition, then Market Structure 2 offers a common initial step in the most feasible of market transition trajectories, and that this could be implemented relatively smoothly in 2009.

5.8 Recommendations

5.8.1 The Consultants recommend that, taking into consideration the broad implications and overall options presented in this Study, Government immediately develops a long-term energy policy strategy for the Hong Kong electricity sector, that: describes the long-term goal for the industry structure; describes the transition path and planned time-line to move from

the current structure to the long-term industry structure; and specifies the relationship between the time-line and external

developments, in particular in the Southern China electricity sector.

Such a long-term strategy should be developed with care. Considerable time is available before the expiry of the current SCAs in which to plan the development of the Hong Kong electricity sector and implement changes. Nevertheless, this time should not be wasted, for it is important to announce the strategy in its broad outline as soon as possible, and later in more detail, well before implementation. This will allow time for the Hong Kong community to respond to the proposed approach and will continue to ensure a stable operating and investment environment for the industry.Selection of near-term resource options should be made with due consideration of their long-term implications, to ensure that they are consistent with the overall long-term strategy and would not preclude the realisation of benefits in the longer-term. Changes in the Southern China electricity sector in the next five to ten years will be key to these decisions regarding resource options.

5.8.2 The Consultants recommend that Government allocates the necessary in-house resources and acquires the appropriate tools within the Administration for the development, detailed planning and implementation of the preferred electricity sector strategy.

5.8.3 The Consultants recommend that, to remove the potential risk to the reliability, a firm plan for overcoming the problem of low frequency oscillations be developed, the timing for implementation be specified — taking into account the combination of the regulatory, logistic and economic considerations described in this report — and the interim solution before implementation agreed. Increasing the capacity of the HEC-CLP interconnection would be a technically sound solution to this problem.(1)

1(?) This recommendation assumes the continued application in Hong Kong of measures — such as Power System Stabilisers (PSS) — that are commonly used to compensate for the effect that high speed Automatic Voltage Regulation (AVR) devices and power control systems tend to


146

5.8.4 The Consultants recommend that the Hong Kong Government establish high level lines of communication with the relevant authorities in Southern China and work with them closely to develop a better understanding and integration of the electricity sector in Hong Kong and Southern China.

5.8.5 The Consultants recommend that a South China Electricity Reliability Council be established. This body should include representation from the interconnected utilities of Southern China and Hong Kong and representative/s from the appropriate Government organisation/s.

5.8.6 The Consultants recommend that moves be made towards implementation of an electricity sector planning process for Hong Kong that is more able to consider least cost resource options for the system as a whole — such as increased use of interconnection between HEC and CLP (as investigated in this Study) as well as increased use of interconnection between Hong Kong and China — than is the case under the present utility planning system that has developed historically based on the model of two separate utilities.

5.8.7 The Consultants recommend that the input parameters used in system planning be reviewed to ensure that they represent the actual system performance to the extent possible, and therefore ensure that the system planned is neither insufficient nor beyond that required.

5.8.8 The Consultants recommend that the generation and transmission planning criteria used in Hong Kong be reviewed with a view to establishing a consistent approach to system planning between the two utilities for the final five-year period of the current SCAs, and also to evaluate the appropriate reliability level for Hong Kong — possibly via a thorough economic analysis of the cost of unserved load for each major group of electricity customers.

5.8.9 The Consultants recommend that the review of generation planning criteria should also explicitly consider the appropriate application of the criteria for planning purposes, including the treatment of the HEC-CLP and the CLP-GPHC interconnections in planning analyses, to ensure that the appropriate level of reliability for Hong Kong can be met in future in the least-cost manner.

5.8.10 The Consultants recommend that a full Level I (generation) and Level II (transmission) reliability analysis of the Hong Kong system, including the interconnection with the Southern China system, be undertaken in co-operation with the two Hong Kong utilities and the appropriate Southern China utilities. This will ensure that the Hong Kong system as a whole — considering its interconnection with the Southern China system — provides the desired level of reliability. This will ensure that the transmission-level reliability is commensurate with generation-level reliability.

have in heavily decreasing damping.


147

5.8.11 The Consultants recommend that, should the long-term electricity sector strategy include a move towards the introduction of competition, the approach to determining the appropriate level of capacity under each market structure adopted in the transition process be carefully considered. If the electricity sector moves to a Single Wholesale Purchaser model, Government may need to take a more active role in determining the planning criteria and the need for new capacity than has been necessary under the existing structure, as part of the process of independently assessing tenders for provision of additional capacity. If the electricity sector moves to a Multiple Wholesale Purchasers structure and eventually to full Wholesale and Retail Competition, the role of Government in determining capacity requirements could be reduced, and market forces allowed to determine them. As explained in the discussion on time frames, this full market arrangement is not likely to be possible in the near or medium-term.

5.8.12 The Consultants recommend that Government integrate the overall long-term strategy for the electricity sector with other broad Government policy areas. Foremost among these are environmental policy (particularly as it relates to air emissions and greenhouse gas emissions from the electricity sector) and the general move towards consideration of sustainable development. Future competitive market integration with generators in Southern China will need to take environmental factors into consideration to prevent competing power stations from allowing higher environmental impacts to gain a wholesale price advantage in a competitive market. Methods exist that are compatible with the operation of competitive wholesale electricity market to achieve the desired outcome, but implementing them will require Government to devote appropriate time and resources to the task and to facilitate the appropriate degree of co-operation with the authorities in Southern China.

Concluding Remarks

5.8.13 The Consultants were required by their Terms of Reference to “provide independent findings and recommendations sufficient to enable the Government to formulate a substantive response to public concern” on the issues of interconnection and electricity sector competition.

5.8.14 This Study has established the initial framework for a more informed debate on the future electricity sector in Hong Kong and the associated negotiations between Government and the two utilities that would be required for restructuring the electricity sector in Hong Kong.

5.8.15 The findings of this Study should be seen as a means of facilitating creative solutions to the public’s demand for improved customer benefits. The utilities should be called upon to rise to this challenge, on the basis of the findings of


148

this Study and exercise their good will, thereby creating a constructive environment in which public debate can take place in the period before the present Scheme of Control Agreements expire.

5.8.16 It is important in this process to impartially examine how public expectations might be realistically met without jeopardising the reliable electricity supply and high level of service that the public in Hong Kong enjoys as a result of the current framework for investment that has been provided for by the Scheme of Control Agreements.


149

6 CASE STUDIES

The following case studies provide some additional information relating to experiences with electricity sector restructuring internationally and some of the key issues that have been encountered, as well as some more general examples relating to market behaviour and competitive markets.


151

CASE STUDIES

CASE STUDY 1: PROJECTING COMPETITIVE MARKET PRICING IN THE USA

Until recently consumers in the United States have had their choice of electricity supplier limited to the utilities franchised to serve their areas. Likewise, electricity suppliers have not been free to pursue customers outside their designated service territories. Utilities have built generation, transmission and distribution capacity only to serve the needs of the customers in their service territories and electricity price has been set administratively, based on the average cost of producing and delivering power, rather than competitively.However, throughout the country there is a move towards making electricity generation markets more competitive. Most States plan to implement significant changes in the procurement and pricing of electricity between now and the first few years of the 21st century.

Observations

In the majority of cases, the prices for transmission and distribution services are expected to continue to be set administratively on the basis of the average cost of service. In contrast, competitive market forces will set generation prices. Buyers and sellers of power will work together, through power pools or one-on-one negotiations, to set the price of electricity.The movement toward competitive pricing of generation has several implications. Generation prices are likely to become more volatile, changing as consumers’ needs fluctuate across the seasons and from hour-to-hour on a hot summer day. For example, as the temperature rises the use of air conditioning will drive up unit electricity prices as plants with higher operating costs are called upon to meet demand. Competitive prices based on marginal costs will be more sensitive to any factors that affect the operating costs of the marginal generators. For example, if the cost of fuel to marginal generators rises unexpectedly, the impact on prices will be readily apparent. Evidence of this is presented in the Case Study 6 on the price spike in the wholesale power markets of the Midwestern USA.Prices to consumers are also expected to fall, at least according to the Government and many knowledgeable industry players. Consider the following examples of price projections under a competitive market by the US Federal Government, and American Gas Association and two consumer advocacy groups.


153

Projection 1: US Department of EnergyThe US Department of Energy recently released its analysis of the Comprehensive Electricity Competition Act. It found major economic and environmental benefits could accrue from introducing competition, including the following: the average family of four would save US $232 annually in the

year 2010, or approximately US $64 billion per year across the nation;

25-40 million metric tonnes of carbon emissions would be avoided annually in 2010; and

residential prices would be 15% lower, while commercial and industrial rates are expected to be 13 and 5% lower, respectively.

Supporting analysis of the Act can be found at: http://www.doe.gov/ceca/ceca.htm.

Projection 2: American Gas AssociationThe AGA commissioned Science Applications International Corporation to undertake a study to estimate the savings to consumers of electricity market restructuring: further market restructuring is expected to promote the use of natural gas for generation. The study estimated prices as shown in the figure below.


154

Projection 3: Consumers Union and the Consumer Federation of America As reported in the Energy Daily (9 July 1998), these two consumer advocacy groups conducted a study on electricity market restructuring and are concerned that ‘efficiency gains from electricity restructuring have been vastly inflated by proponents and residential customers will see higher-not lower-rates unless policymakers establish strong safeguards against the “likely” abuse of market power.’

The study noted the following: utilities faced substantial transaction costs, such as new system

management and marketing expenses, that would raise consumer rates by 12 to 22%;

costs would be shifted from the most attractive customers for utilities (ie large industrial customers) to residential customers, raising householders’ rates by 10-20%;

given past monopoly conditions, electricity markets are likely to be plagued by the abuse of market power, leading to price hikes of 9-23%; and

stranded cost recovery would be heavily allocated to residential customers, resulting in cost increases ranging from 7 to 10%.

Why Such Differences?

Many of the differences in the projections of savings from market restructuring can probably be accounted for by examining who did the study. The government may have an interest in seeing that its policies are vindicated while the AGA most certainly wants to promote the sales natural gas. Determining the bias of the consumer groups may be more difficult, but traditionally they have been reluctant to endorse changes in regulatory structures that will make it harder to ensure consumer protections are in force. If only one utility serves a particular geographical area, it will be much easier to monitor for consumer protection.However, it is most likely that the US Government report is the most objective and was certainly the most extensive. Consumer groups have also been notoriously wrong in the past with their underestimates of the consumers savings gained through the restructuring on both the telecommunications and natural gas markets in the USA.Lessons for the Electricity Industry in Hong KongWhere market forces can be utilised to place pressure on competitors they are likely to reduce prices. The issues of concern to Hong Kong are in many ways very similar to those encountered elsewhere, eg consumer confusion over choice, potentially large stranded costs and market dislocations as firms have to change their behaviour.


155

However, as this study does not recommend that Hong Kong move towards a competitive market until 2008 due to the lack of competitors (eg until the Mainland China market opens), many of these concerns can be addressed at a later date after more countries in the world have experience dealing with them.


156

CASE STUDY 2:INTERNATIONAL EXPERIENCE OF ELECTRICITY COMPETITION — USA, UK, BRAZIL AND THE EU

United States of America

Change in the electric utility industry in the USA is occurring at a rapid pace, and a new type of organisation, the power marketer, has evolved to play a key role. Up to now, vertically integrated traditional electric utilities (Model 1 as described in the main report), which had generation, transmission and distribution systems, were the only entities which bought and sold electricity in bulk. Now, power marketers, which usually do not own any of the traditional assets associated with electric utilities, have emerged as major players in the competitive market.Although the wholesale market for electricity is becoming increasingly competitive, that does not mean that it is unregulated. Power marketers whose rates are on file with the Federal Energy Regulatory Commission (FERC) are considered to be public utilities under the Federal Power Act and must comply with a number of regulations which apply to all public utilities. However, many of the regulations which customarily apply to traditional public utilities have been waived or relaxed for power marketers, eg they are permitted to charge market-based rather administratively set rates. If the power marketer or its affiliates owns or controls electricity generation or transmission facilities, it will be required to perform a market dominance study which demonstrates that the power marketer and its affiliates have no market power in the relevant markets.The USA is acting on the generally held belief that competitive electricity markets will lower prices to residents and businesses, and thus many States are now addressing electricity market restructuring. However, their progress varies significantly.

California

In California, where plans are furthest developed, the Public Utility Commission (PUC) and State Legislators have established a programme to provide competitive electricity markets and consumer choice in 1998. In principle, the plan calls for full recovery of utilities’ stranded costs over a ten year period through a ‘non-bypassable competitive transition charge’(1). To ensure that small customers will benefit from competition, the Californian plan calls for rates to be cut by no less than 10% from 1 January 1998 and 20% by 1 January 2002. With respect to transmission services, zonal rates based on the cost of services are being set in accordance with FERC’s Order 888 (see Case Study 4 on

1(?) Non-bypassable means that the charge will be imposed in such a way that consumers cannot avoid paying it, whether they stay with their current utility or choose a new supplier.


157

stranded costs for further discussion). Prices will be differentiated by geographic locations within the State, because transmission capacity between zones is limited. Subsequently, market clearing prices will differ between zones. Distribution services are to be priced according to a Performance Based Regulation mechanism (PBR). In a typical PBR system, prices are linked to a key economic index adjusted for expected increases in productivity; a comparison of such mechanisms and their applicability in Hong Kong can be found in Section 3.3.27 in the main report.To ensure that the electricity market in California will operate competitively, the plan calls for utilities to functionally unbundle their operations. In other words, they must run their transmission, distribution and generation operations as if they were separate companies. Operation of the transmission grid and the dispatching of generating plants to meet demand are to be performed by a new, non-profit institution — the Independent System Operator (ISO). The transmission facilities will still be owned by the utilities, but their operation will be controlled by the ISO. A separate non-profit institution — the Power Exchange (PX) — will be formed to provide an efficient, competitive auction open to all suppliers. The purpose of the PX will be to facilitate a short-term pool for electricity transactions.

United Kingdom

The competitive electricity market in the UK consists of three large generating companies (plus a number of small Independent Power Producers, or IPPs) that produce electricity, and 12 privately owned regional electricity companies (known as RECs) that distribute the power to consumers. The transmission facilities are owned by the National Grid, which operates a large power pool and runs a spot market(1). Experience to date suggests that the splitting of generation and transmission does not adversely affect reliability. In fact, in the UK, Independent Power Producers are constructing new power plants without long-term contracts. Due to the increased incentive for electricity generators to bargain with equipment suppliers, the price of combined-cycle gas-fired power plants has also fallen by 30%. There have also been substantial improvements in how efficiently generating facilities are run, including a 30% increase from nuclear power plants. However, total employment has been reduced by more than 40% between 1990 and 1994.Although the costs of electricity generation have been lowered in the UK, the cost reductions have largely been retained by the 1(?) The spot market operates as follows: each afternoon, the generators submit bids to supply a given amount of power to a given price for each 30 minutes of the next day. The marginal cost for the entire system for a given 30 minutes is essentially the bid price of the last unit dispatched. The spot price is the sum of the marginal cost and a reliability component, based on the estimates of the probability of an outage and consumer’s losses if an outage occurs. If the bid is accepted, the bidder receives the spot price.


158

generation companies in the form of higher profits rather than passed through to consumers in the form of lower prices. The evidence to date suggests that competition in the UK has had only a limited effect on retail electricity prices. This is mainly due to the fact that the market is dominated by only three large power utilities, with only a small number of IPPs, and the competitive price of electricity to consumers remains much higher than the incremental cost of production. This is an indication that market power exists. The UK experience provides an excellent illustration that simply allowing competition does not guarantee that a sufficient number of players will enter the market and that the market power of producers will be eliminated, thus maximising benefits to consumers from lower electricity prices. The UK Government is currently embarked on a review with a view to removing market power such that consumers will gain more benefit from electricity competition.

Brazil

Over the last three years, the Brazilian electricity sector has experienced fundamental changes in its regulatory and commercial structure. Largely due to severe fiscal constraints that sharply reduced the government’s ability to continue acting as the leading investment source in the power sector, privatisation of electric utilities is currently considered a government priority. Aiming to attract domestic and international capital to the power sector through privatisation, the Brazilian Government decided to revise the current regulations and formulate a framework more appropriate to a competitive and global market.Despite an increase of private capital in the sector for the past two years, federal and local state governments still control most of the electric utilities in Brazil. Even after the privatisation of seven power distribution companies and one small generation company, the government still participates with over 80% of the total electricity sales in the country. By the end of 1998, an additional 14 companies are scheduled to be privatised which would reduce the government’s market share to approximately 60%.The proposed model for restructuring the Brazilian electricity sector focuses on a set of regulatory measures to facilitate competition essentially by: separation of the activities of generation, transmission and

distribution/retail; provision of open access to transmission and distribution

networks; and introduction of a competitive wholesale electricity market for

generation.


159

The structural implications of the proposed model and determination of the tariffs would result in the following: Generation. With the objective of increasing competition in

generation, the model suggests restrictions on vertical integration by imposing a limit on the level of sales of generation from the company to its own distribution arm. The tariffs would be determined mostly by freely negotiated long-term contracts and the remaining energy would be sold in the spot wholesale market.

Transmission. In order to achieve open access to the transmission system, generation and transmission would operate independently. Transmission assets would be separated from generation assets both at the federal level and at the local state level. There would be at least two transmission companies at the federal level to avoid a single national company contesting the authority of the independent operator. The companies at the local state level would have to create separate wholly owned subsidiaries for their transmission assets, or offer their transmission assets to one of the federal companies in exchange for shares in the federal transmission company.

Distribution and retail. Most of the distribution and retail activity would not require immediate restructuring. There are already a large number of companies and more should be created through the division of some of the largest existing distribution and retail companies. It is envisaged that distribution charges will be based on open access to the distribution network as a way to enhance the competitiveness and efficiency of the distribution system.

The figure below illustrates the 15 year transition period in which the existing energy traded on the basis of initial contracts would gradually and increasingly be traded in the free market. The length of the transition period is one of the main items being reviewed and the government had indicated that it might opt for a shorter amount of time.


160

Electricity Market Transition Period in Brazil

SPOT market

Competitive market - freelynegotiated bilateral contracts

Initial contratcsphase-out tomarket rules

Initial contracts atcurrent prices

- avoid tariff shocks- security to investors

TWh

Year “6” (maximum) Year “15” (maximum)

10-15

%85

-90%

Curre

nt si

tuati

on

Source: Global Utilities Finance Report 1998

The European Union

On 19 December 1996, after almost five years of debate on the text and even longer on the principle, the Council for the European Union finally adopted Directive 96/92/EC on the liberalisation of the electricity industry. The Directive envisages greater competition in the generation and supply of electricity and also provides for some unbundling of the various aspects of electricity supply activity.The member states of the European Community can follow two different competitive models in generation: Authorisation. A free market entry scheme for new generation,

subject only to an objective consent or authorisation process. There would be no central generation planning.

Tendering. The second model envisages central planning but no generation monopoly for the incumbent utility(ies). New generation demand would be put to competitive tender in which it would be open to the incumbent utility to participate. Existing capacity from other interconnected systems would also be free to compete.

There would again be two main different structures in supply: Negotiated third party access (called Multiple Wholesale

Purchasers within the Hong Kong study). This permits bilateral trading relationships to occur between generator and consumer using access to the transmission and distribution networks.


161

A single buyer system (called Single Wholesale Purchaser within the Hong Kong study). This is an entity with a pivotal role in an electricity system with both a purchase and supply monopoly, ie it is the sole purchaser of generation within the relevant area and therefore all consumers must purchase from it.

Another feature of the Directive is unbundling, in that the functions of the transmission system operator and/or single buyer are required to be operationally separate from those of generation and distribution. Furthermore, there are requirements that account for the generation, transmission and distribution aspects of the business to be unbundled to prevent cross-subsidy of activities.The possible impact of the Directive is seen to have been diluted over the long years of debate and compromise. This is particularly so with the absence of an access right for distributors, the progressive opening up of only about one-third of the Community market over the Directive’s timetable and the limited level of unbundling. Also, the public service obligation (PSO) concept — that incumbent utilities are legally obliged to supply all customers in their area, even if they are not commercially attractive — and the extensive opportunities for disapplications, derogations and exceptions all give reason for fear that implementation will be contentious. There is little doubt, however, that an impact is already being felt(1), for example: The advent of greater competition in generation, even in the form

of a tendering system, will provide a major challenge to incumbent utilities. It remains to be seen whether preferences towards incumbents survive in practice.

In the case of competition in supply, a similar stranded investment problem emerges, particularly if a local distributor has long term purchase commitments to relatively high cost power. Again, such a distributor could simply be bypassed.

Attempts to protect home markets while attacking those in other member states — known as ‘negative reciprocity’ — will be a highly contentious issue and may well end in dispute. Given the ambiguous nature of the relevant provision in the Directive and the residual powers of the Commission under the EC competition rules, disputes which are allowed to arise will be complex.

Although less adventurous than the Commission and some member states would have liked, the Directive represents a move towards generalised liberalisation, unthinkable even six years ago. Now the question is not whether or not there should liberalisation, but rather how much and how soon is should occur.

1(?) Electricity Liberalisation in Europe, Cameron McKenna, 1997.


162

CASE STUDY 3: HOW MANY PLAYERS ARE NEEDED FOR A MARKET TO BE GENUINELY COMPETITIVE?

Economic Theory and What it Tells Us

Economic theory often provides clear and reliable pointers to the behaviour of firms under conditions of monopoly or free competition. In markets which are genuinely free (such as many financial markets) or which are true monopolies (for example for unique pharmaceutical products which are under patent protection), the predictions of economic theory explain crucial aspects of firms’ behaviour. Given certain key information about a market, it is generally straightforward to identify what sorts of policies a firm will adopt in the market. However, few markets correspond precisely to the characteristics of either perfect competition or monopoly. For example, the vast majority of monopolies in advanced economies today are tightly regulated. With regard to free markets, there are even more differences between the theoretical model and reality, particularly in newly liberalised markets. Differences include:Free markets require zero barriers to entry. In practice, all markets have some barriers to entry. For example, consumer goods manufacturers often invest heavily in developing brand loyalty, which a new entrant will find hard to emulate. Barriers to entry are even more pronounced in utilities, where the heavy capital expenditure required is often a major obstacle to new entrants.Free markets also require perfect information. This condition requires that all players have access to the same data about the market. In many markets, off-the-shelf market research can help to meet this condition. But, particularly in the case of utilities, this condition is often not met (often due as much to data protection laws as to commercial secrecy). The advantage incumbents have in a particular geographical location is viewed by many as the impetus behind the growing trend for privatised UK utilities to take over other utilities in different sectors but the same region(1) eg a power company buying its overlapping water provider.Finally, and perhaps most importantly for the regulation of utilities, free markets require that no one firm or group of firms can influence market prices. In many markets this condition is clearly not met. Even with elaborate and sophisticated regulation, there are numerous examples of ingenuity on the part of dominant players in order to exert market power. Where the conditions required for the theories to be valid do not exist, theoretical insights into company behaviour may be of limited use to the regulator.

1(?) See for example the Financial Times, 18 July 1998.


163

Lessons for the Electricity Sector in Hong Kong

It can be seen that economic theory alone is of little use when attempting to predict the behaviour of firms in markets which do not conform to free market models(1). A central question is therefore:What conditions need to exist before firms begin at act as they would in a free market thus bringing the benefits to consumers of open competition?In the case of electricity de-regulation, vigilance by the regulator can solve many of the key problems. For example, barriers to entry can be reduced greatly by non-discriminatory common carrier regulations, which allow all generators access to transmission systems on equal terms. Regulators can also attempt to ensure that no single firm or group of firms can manipulate the market. However, experience around the world has indicated that this is difficult when there are only a small number of firms in the market.The UK has the longest experience of deregulated electricity markets. This is particularly true in the generation sector, where power producers were split up into three companies at the time of privatisation in the early 1990s. The motives for splitting the then Central Electricity Generating Board into only three companies were varied(2), however, the end result has been that substantial abuses of market power were identified and major structural reform in the market is now being proposed. Investigations by the electricity regulator in the UK recently concluded that the two big non-nuclear generators (National Power and PowerGen) engaged in anti-competitive behaviour and were able to manipulate the wholesale electricity market. In addition to reforms to the wholesale market, the regulator has recommended that the two generators be forced to divest themselves of a significant proportion of their high quality, base-load generating capacity. The intention is to break up the market for generators and create the conditions for more effective competition(3). The US experience of market deregulation is much shorter, but there are already complaints by regulators that in regional markets dominated by two or three firms, dominant players are colluding to prevent future competition. In particular, complaints have already been made in Rhode Island(4) and Pennsylvania(5). More time will

1(?) For more information on the theoretical perspective see, for example, Richard G Lipsey, An Introduction to Positive Economics, Fourth Edition, Chapters 23 and 24. 2(?) Principal considerations were (i) the desire to privatise nuclear power stations, resulting in a need to create one company (which became National Power) with the financial muscle to pay for decommissioning of old nuclear plant, and (ii) the desire to obtain a high sale price and ensure good market performance upon privatisation. 3(?) Conclusions of the Review of Energy Sources for Power Generation and Government Response to fourth and fifth Reports of Trade and Industry Committee, UK Department of Trade and Industry, October 1998. Available on the Internet at http://www.dti.gov.uk/energy/index.htm. 4(?) Providence Journal-Bulletin, 12 May 1998. 5(?) The Patriot [Newspaper], Harrisburg, Pennsylvania, 3 October 1998.


164

be needed to validate these early concerns (competition was only introduced in January 1998 to these areas), but a trend recognisable from the UK experience may well be beginning.

Comparison with Other Sectors

The power sector is not alone in needing careful supervision to prevent abuses of market power by dominant players. Set out below are a number of examples of various forms of market manipulation experienced when only a few players have been able to dominate their markets:

SupermarketsBritain’s food retailing industry has been increasingly dominated by major supermarkets in recent years. Four-fifths of all food sold is now retailed through supermarkets, and four firms dominate this share. In recent years supermarket chains have been accused of indulging in a number of anti-competitive practices including: colluding to use buying power to drive down supplier prices

without passing on gains to customers; agreeing to permit local monopolies for supermarket services;

and putting pressure on suppliers to prevent new entrants to the

market.The supermarkets dispute these allegations. However, despite the fact that UK farmers are among Europe’s most efficient, that UK labour costs are considerably lower than those of most EU nations, and that UK supermarkets have been allowed to expand in cheap, out of town centre locations, an investigation by The Sunday Times newspaper found food prices to be considerably higher in Britain than elsewhere in the EU(1). The paper found that a basket of good that would cost 82.05 (about HK$1 066) in the UK could be purchased for 60.20 in Italy, 60.14 in France, 53.31 in Germany and 49.55 in Holland.

Steam Turbine Generator Manufacturers In the 1950s and 1960s, 95% of all steam turbine generators made in the USA were built by just three companies. Until 1958 prices for their products were kept high by collusive agreements between the three firms. Only when the Tennessee Valley Authority threatened to purchase steam turbines from foreign manufacturers, and a Federal anti-trust suit was rumoured, did prices fall. In particular, commentators considered that the opening up of the market to

1(?) The Sunday Times, 27 September 1998.


165

more (foreign) competition was the main reason for the subsequent vigorous price competition(1).

CigarettesThe US tobacco industry has one of the highest levels of ownership concentration of any manufacturing industry. The industry has long been dominated by just three or four firms (including RJR Nabisco, Philip Morris and Brown & Williamson). Since the 1930s, anti-competitive behaviour has been noted, as evidenced by very low competition for tobacco crops (even in times of shortage) and profit levels which are well above long-run averages in manufacturing industries(2). Industry observers consider that this situation persists to this day(3).

TelecommunicationsThe 1980s and 1990s have seen a massive wave of deregulation activity affecting almost all aspects of the telecoms industry. The US was an early pioneer in this field, with the break-up of the AT&T monopoly — known as ‘Ma Bell’ — into seven regional ‘Baby Bells’. In order to stimulate competition in a market where one firm had been dominant, local Baby Bells were prevented from offering long distance telephony until their markets were sufficiently open and competitive for local telephone services. New local competitors were permitted to offer both long distance and local telephony but they require third party access to the networks of the Baby Bells. Despite the break-up of AT&T in 1984, only now have some Baby Bells been given the right to sell long distance telephony(4), giving a graphic illustration of the difficulties in opening up markets which have long established incumbents. In Hong Kong, the Office of the Telecommunications Authority has also adopted aggressive competition rules, particularly in the case of mobile telephony (see Case Study 5 for more details on telecommunications regulation in the SAR).

1(?) Richard G Lipsey, ibid. 2(?) See Richard G Lipsey, ibid. 3(?) For example, see the projection by Wall Street stockbrokers Brown Brother Harrimen who reported that "the nation's strongest price oligopoly will be sustained." Investext Broker Reports, 17 August 1998. 4(?) For a recent example see the case of Ameritech and US West, two Baby Bells who were recently denied the right to even sell the long distance services of independent provider Qwest. Newsbytes News Network, 29 September 1998.


166

Implications for the Electricity Sector in Hong Kong

These examples indicate that opening up markets to competition, and preventing anti-competitive behaviour from emerging in previously open markets, is a significant challenge for regulators. It is also notable that the examples above come from countries with a strong history of consumer protection and competition policy. This in itself is indicative of the need for aggressive, pro-active protection of consumers’ rights. The examples cited above, and many other similar cases, also demonstrate that simply allowing the right conditions for competition does not guarantee that competition will ensue. In particular, where the number of firms dominating an industry is small, the chances of effective competition without an aggressive attitude towards the protection of consumers interests are smaller.The number of competitors seems to be a crucial part of this function, although where industry wide(1) economies of scale exist, there is often a case for establishing a limited number of strong competitors rather than a large number of weaker ones. This is particularly the case in the transition period from monopoly to competition where levels of service and corporate viability are critical. More competitors are usually, but not necessarily, better and individual circumstances must be taken into consideration.

Local Market Dominance on Hong Kong Island

As discussed in Case Study 2 and elsewhere within the main report, for competition to be effective, no individual company should be able to exert control over the market such that it can dictate prices to be paid in the market. In practise it is nearly impossible to completely eliminate control over prices, but it should be minimised as far as possible.The issue of market dominance on Hong Kong Island is of particular concern due to the following: access to a limited capacity interconnector; only one power station currently supplying Hong Kong Island; and even if interconnection was increased, it is unlikely to be

economic to increase it to the point that it could supply the entire Island.

Assuming that the Hong Kong market was open to competition and the cross harbour interconnectors were not large enough to supply the entire load of Hong Kong Island, the owners of the Lamma Power Station would be in a position to exert influence such that they could determine the price to be paid by consumers (see the figure below).

1(?) Industry wide is contrasted to plant level in this situation. Electricity generation plants are not considered by the Consultants to have significant economies of scale whereas the industry as a whole may.


167

Schematic Representation of Energy Supply Scenarios

Now(verticallyintegrated)

Future(competitive

market)

Percentof GWh

Consumed Allconsumers

pay thesame price Competitively

SuppliedAcross

Interconnector

Captive ofLammaPowerStation

Assuming that the interconnector to Kowloon was increased to 1 000MW and total load was 3 000MW, it could be argued that the first 1 000MW of power sold across the interconnector would be competitive (assuming that the market in Kowloon was competitive). If the Lamma Station owner was intent on maximising its profits, it could ignore the bidding for the first 1 000MW and then charge whatever the market would bear for the remaining 2 000MW. Similar to other companies in other parts of the world who have tried to exercise market power, the Lamma Station owner would of course be subject to certain political and economic pressures. However, as would any firm, it should be expected to maximise its profits. This situation is very similar to that which was the impetus for the review that has recently been completed on the UK’s electricity market (see Case Study 2). There it was found that market power, although explained away by the incumbent generators as maintenance scheduling and other issues, was being exercised by the three large power producers. Also, consider the following article that describes the situation in New York City, a very close corollary to that occurring on Hong Kong Island (Con Edison is the utility that serves New York City):‘Con Edison is in the process of selling off much of its generating capacity, and the income derived will be used to reduce stranded costs. If Con Edison achieves a windfall on the sale of its plants, the result could ultimately be lower prices for the customer. In New England, the sale of generating plants for substantial sums increased savings for consumers by up to 10%. The players who have entered the New York City market thus far seem to be conservative in their spending, as evidenced by last year's capacity auction, where the price the bidders paid was far less than


168

expected. However, because New York City is a load pocket (ie, due to transmission restraints during peak loads, up to 70% of the power used must be generated in-city), potential buyers are likely to see additional value in the plants and bid up the prices.

‘Not only will the sales price of the plants have an impact on the cost of electricity, the fact that the plants will have new owners will also have an impact. As previously mentioned, New York City is a load pocket. Therefore, every player in this market must purchase some in-city capacity. For Phase I, that meant participating in an auction to purchase capacity from Con Edison. After the generating plants are sold, that means purchasing capacity from one of the new owners. The cost of capacity may ultimately be forced down by competition between the new owners. Or the prices may be forced up by new players intent on entering this market.’Source: Diana Sweeney, Associate Director, Insignia/ESG Energy Services writing in Real Estate Weekly, Vol.45, No.7.

Lessons for the Electricity Industry in Hong Kong

If market power is identified on Hong Kong Island (or elsewhere within Hong Kong) and a competitive market is, or will be implemented, Government may grant authority to the owner of Lamma to charge market-based rates if it is comfortable that it can mitigate market power through any of several measures, including: utility generation divestiture; on-going market study updates; specific open-access tariff requirements; allowing market-based rates subject to a revenue cap or a bidding

rule; and cost-based recourse rates or price caps, pending divestiture.(1) If appropriate mitigation cannot be developed, Government may want to consider other options such as traditional cost-of-service rates or flexible pricing constrained by a cost-based rate cap that does not exceed the rate per unit developed from the traditional overall revenue requirement.(2)

1(?) This is similar to the issues addressed by FERC for Pacific Gas & Electric Co, 77 FERC ¶61 265 at 62 083, 62 087.2(?) Ibid. at 62 087.


169

CASE STUDY 4: STRANDED COSTS

Introduction

The recovery of stranded costs is one of the most contentious issues confronting regulators in promoting competition. Stranded costs (or assets) are costs that have been prudently incurred, as approved by governments, by utilities to serve their customers, but which cannot be recovered when the market or regulatory structure changes.Stranded costs are a particular challenge for privatised electricity industries as the companies are often traded on stock markets and their valuation is determined through the market’s expectations. To both maintain market confidence in the sector and to honour Government contracts with the utilities, there may be grounds to allow the utilities to collect costs they incurred prudently.During the transition to a competitive environment in the USA, the Federal Energy Regulatory Commission (FERC) noted that some utilities may incur stranded costs as wholesale customers leave to purchase power from alternative sources. Accordingly, Order 888 provided a mechanism for recovery of stranded costs with a view to ensuring an orderly and structured transition to a wholesale market that would increasingly rely on market-based rates in the future.Stranded costs may be an issue for Hong Kong, especially for contracts already approved by Government and which contain commitments past 2008.

Order 888In Order 888, FERC reaffirmed ‘that the recovery of legitimate, prudent and verifiable stranded costs should be allowed’. FERC’s directive is grounded in the belief that the recovery of stranded costs ‘is critical to the successful transition of the electricity industry to a competitive, open access environment’.Within Order 888, FERC aims to balance a number of critical interests to achieve a fair and orderly transition to competition. These include sustaining the financial stability of the industry, upholding the regulatory bargain under which large investments were made by the industry in the past and not shifting costs to customers that had no responsibility in causing stranded costs to emerge. The FERC acknowledged that stranded cost recovery may delay some of the benefits of competition, but concluded that customers will benefit in the long run from a fair and orderly transition


171

Definition and Calculation of Stranded Costs

The FERC definition of a wholesale stranded cost is : ‘any legitimate, prudent and verifiable cost incurred by a utility to provide service to a wholesale requirements customer, a retail customer, or a newly created wholesale power sales customer that subsequently becomes, in whole or in part, an unbundled wholesale transmission services customer of such utility’.

During the transition period prices are to be determined by market forces, and high-cost utilities may be unable to fully recoup the embedded costs of their investments. The amount by which the embedded cost of a utility exceeds the market value of an asset is generally referred to as stranded cost.Order 888 states that direct assignment of stranded costs computed on a revenues lost basis is the appropriate method for recovery.

Treatment of Stranded Costs

A report prepared by the Energy Information Administration of the US Department of Energy(1) highlights some of the views held by the different market participants on the issue of stranded costs: The Edison Electric Institute, an association of utilities, maintains

that a significant portion of the stranded costs is attributable to public policies of the past. Recovery of stranded costs is thus based on equity and fairness arguments.

Investor-owned utilities hold similar views and claim that they should be entitled to a full and timely recovery of stranded costs. They claim that under competition, recovery of stranded costs is deemed necessary to maintain the financial viability and operational reliability of this critically important infrastructure.

Industrial users are particularly interested in securing lower electricity rates, and they view full stranded cost recovery as an impediment delaying the benefits of competition. Some commercial and residential customers do not support full stranded cost recovery for similar reasons.

Estimation of Stranded Costs

Arguments supporting recovery of stranded costs in the USA vary, with estimates of nation-wide stranded costs of between US$100 and $200 billion(2).

1(?) The Changing Structure of the Electric Power Industry: Selected Issues, 1998, Energy Information Administration, US Department of Energy2(?) The Changing Structure of the Electric Power Industry: Selected Issues, 1998, Energy Information Administration, US Department of Energy


172

At least one market analyst(1) has published estimates of the stranded cost compensation that may be required if the Scheme of Control Agreements were prematurely terminated in 1999 and increased interconnection and a power pool introduced. The details of the analysis and the full set of assumptions underlying the numbers are not provided in the source document and therefore, the numbers are not quoted in this report.Compensation may be applicable should the SCA be re-negotiated to the detriment of the utility companies before they are due to expire in 2008. In the event of a general restructuring of the industry where the utilities would be required to assume prime responsibility for performance, unit demand and price risk, it could be argued that such changes would warrant grounds for payment of compensation.


Potential examples of stranded costs in Hong Kong include: the gas pipeline contract for CLP’s Black Point Power Station that

expires at the end of 2015; CLP’s contract to purchase power from Daya Bay nuclear power

station until 2014; HEC’s possible LNG contract for its proposed Lamma Extension

should the contract extend beyond 2008; and any other assets incurred to serve load forecasts accepted by

Government, but which are uneconomic in a changed regulatory structure.

The experience with stranded costs internationally is varied and contentious, but most countries are taking the view that at least some portion of prudently incurred stranded costs should be recovered by utility investors.

1(?) Indosuez WI Carr Securities, Hong Kong Research — Utilities Sector — The Future: Working With China, August 1998, page 14 and ING Barings Hong Kong Research: Power Sector Update 18/9/98.


173

CASE STUDY 5: COMPETITION IN THE TELECOMMUNICATIONS INDUSTRY IN HONG KONG

Introduction

There has been a world-wide trend for competition to find its place in the market for telecommunications, including Hong Kong. In June 1995, a total of four fixed telecommunications network service (FTNS) licences were issued to the Hong Kong Telephone Company Limited (HKTC) and three new entrants.

Tariff Rules and Regulatory Approach

Significant tariffing rules have been imposed on HKTC by the Telecommunications Authority (TA) to prevent anti-competitive actions in view of HKTC’s still dominant market position. It has not however been the intention of the TA to set in stone these tariffing measures. The ultimate objective is to abolish all tariff controls once a truly competitive market has been achieved.All of the four FTNS licences issued in June 1995 included tariffing rules, however since then the TA has waived the majority of them as it became clear that the licensee was not in a dominant position. However, they have not been waived for HKTC as it is foreseen that they are a dominant player and are likely to remain so for some time.In managing the transition from a monopolistic environment to a freely competitive market, the TA perceives its role is primarily to establish suitable conditions for the industry to operate with the minimum necessary level of intervention. This involves the setting up of suitable regulatory apparatus and sufficient competitive safeguards such that consumers’ interests are protected and that new operators can enter the market with fair opportunity to compete. This minimum intervention regulatory approach is seen to have the advantage of smoothing the transition to a truly free and open competitive market as efficient entrants succeed on their own merits. It also has the added benefit of avoiding sub-optimal decisions being made by an overly intrusive regulator.

Competitive Checklist

The tariffing relaxation steps were tied to a set of observable, measurable and objective criteria in an effort to provide some certainty in the process, and to also focus on the contributions necessary from HKTC to ensure the operation of a competitive environment. The agreed checklist is detailed below.


175

Hong Kong Competitive Checklist for Telecoms

Interconnection Forms

Type I interconnection is offered by all FTNS licensees on non discriminatory commercial terms and conditions to all other FTNS licensees.Type II interconnection is offered by HKTC on non discriminatory commercial terms and conditions to all other FTNS licensees.

Interconnection Charges

Interconnection charges are based on reasonable relevant costs including cost of capital.

Sharing of Facilities and Co-location of Equipment

HKTC offers sharing of its facilities including co-location of other FTNS licensees’ equipment on non discriminatory commercial terms and conditions.

Calling Line Identification

CLI is offered by all FTNS licensees to other FTNS licensees on a ubiquitous basis.

Printed directory and telephone directory service

Printed directory offered by any FTNS licensee must list all FTNS customers. Telephonic directory service must utilise a unified database covering all FTNS customers.

IDD access

‘00X’ codes are assigned to FTNS licensees for customer access to the international gateway.

Number portability

Numbers used for fixed telephone services, personal numbering service and ‘800’ freephone service are portable between operators via an intelligent network system and are subject to commercial terms and conditions.

Resale

HKTC provides its services to resellers and other FTNS licensees at HKTC’s published retail rates and possibly at less than the standard retail rates for FTNS licensees.Source: Enforcement of the FTNS Tariffing Rules in a Developing Competitive Environment, Statement by A Arena, Telecommunications Authority, Hong Kong.


176

The TA will continue to monitor the development in the industry against the competitive checklist and once all of the conditions have been met, the TA proposes to implement a streamlined process. There has been a general concern that adequate procedures should be undertaken with regards to HKTI’s (Hong Kong Telecom International Limited) monopolistic power and also HKTC’s dominant position. Subsequently, deliberate regulatory steps were taken to separate HKTI’s and HKTC’s operations and they are now separate entities. The new entrants also requested that a threshold be placed on market share, however this was revoked by the TA because it is seen that market share is a reflection of customer inertia and brand recognition, thus it would be inappropriate to be one of the competitive conditions, as this would limit the role of the market in encouraging competition.

Competitive Benchmarks

Every tariff application is screened against certain competitive benchmarks defined in the focus areas of price level, price discrimination and product packaging. Following this screening process, applications are divided into two streams, one being fast stream (Stream I) and the other involving detailed review (Stream II). Generally, a positive answer to any one of the three benchmarks immediately involves a detailed review, as illustrated below.


177

Is the service(s) in the tariff application priced below the fully

distributed cost?

No Yes

Does the tariff application involve price discrimination betweendifferent types of customers?

No Yes

Does the tariff application package one or more services together?

No Yes

Stream I

Fast track

Stream II

Detailed review

Is the service(s) in the tariff application priced below the fully

distributed cost?

No Yes

Does the tariff application involve price discrimination betweendifferent types of customers?

No Yes

Does the tariff application package one or more services together?

No Yes

Stream I

Fast track

Stream II

Detailed review

It is undeniable that significant progress has been made on implementing fair competitive conditions in the Hong Kong FTNS market. Such progress has ensured that there was immediate streamlining of the tariff approval process, thus saving resources in preparing and vetting tariff applications. To ensure that the proposed simplified procedures would not undermine the government’s policy of promoting fair competition, the TA has considered conducting some post-implementation audits if necessary.

Complaints Relating To Dominant Market Position

There have however been complaints lodged with the TA with regards to the marketing practises of HKTC. HKTC have been bundling the sale of equipment with the offer of free telephone services. TA has subsequently undertaken an investigation into the


178

effect of the marketing practices on the competition in the market for the supply of local telephone services within Hong Kong.The TA’s investigation established that HKTC entered into an agreement or an arrangement with an affiliated company whereby HKTC supplied telephone services to the purchasers of bundled packages marketed by the affiliated company. HKTC would bill the affiliated company, instead of the customers, for the telephone service supplied during a specified period after the sale of the packages. The affiliated company would pay HKTC the charges for the telephone service provided to the customers at the fully tariffed rate. After the initial period, HKTC would bill the customers directly for the telephone service supplied. Thus the purchasers did not have to pay the tariffed charges for the telephone service received during the initial period. This was the inducement to the purchasers to use the telephone service of HKTC, and not those of other FTNS operators. The bundling deprived the other FTNS operators of the opportunity of marketing their telephone services to purchasers of the packages. TA considered that the bundling therefore placed the other FTNS operators at a significant competitive disadvantage in the supply of telephone services to the purchasers of the packages. This arrangement was a circumvention of HKTC’s FTNS Licence.In the cases identified, the customers were offered prices during the initial period which would be predatory and anti-competitive if offered by HKTC, as they were below the relevant costs over the free service period. Whether the price was predatory or anti-competitive should be considered from the point of view of the customer because of the price payable by the customer affected his/her decision in the choice of supplier. Even though HKTC might have received payment from its affiliated company for the fully tariffed price of the service, this did not affect the perception of the customer that the service was free during the initial period. TA considered that if the free service had predatory and anti-competitive effect, it would still have this effect irrespective of whether HKTC was receiving the fully tariffed payment from its affiliated company.Although the cases might not individually constitute a substantial restriction on competition, TA considered that they form a pattern of behaviour involving a network of arrangements or agreements which should be considered in totality by the TA to assess the purpose or effect on competition. The arrangement/agreement in question enabled HKTC to effectively circumvent its licence obligation. As HKTC was and still is the dominant operator in the supply of telephone services bundled with the equipment sale, the arrangement/agreement had the effect of shoring up the dominance of HKTC. This meant that HKTC had the ability to reinforce its market power in the provision of local telephone service and to create a hindrance to new FTNS operators attempting to compete in the market. As such, the TA was of the opinion that the network of arrangements/agreements as a whole constituted a conduct which


179

was having the purpose or effect of significantly restricting competition in the market for the supply of the telephone service in question.TA subsequently advised HKTC to stop the infringement and to comply with the revoked licence conditions, by suspending the practises and refraining from engaging in similar practices in the future.

Relevance to Electricity Industry

Although the telecommunications industry is different in many ways from the electricity industry, there are also many similarities. For example: non-discriminatory access to networks is critical for the success of

restructuring; many of the activities of both were perceived to be natural

monopolies until fairly recently; even with the introduction of new industry players, the existing

companies will most likely remain the dominant players; and many consider the goal for each industry as having as little

government involvement as is necessary to ensure open competition.


The highly successful experience in Hong Kong of the TA in facilitating competition can offer a number of lessons for Hong Kong’s electricity industry, namely: prior to any kind of market restructuring, it is difficult to predict

how the market players will react; dominant player regulations should be flexible and should not try

to predict behaviour, but should provide the regulators with the ability to identify and correct what will most likely be very dynamic market dominance behaviour; and

the regulators should have as their objective withdrawal from price setting and profit regulation once certain competitive market conditions are met.


180

CASE STUDY 6: PRICE SHOCKS IN THE US WHOLESALE MARKET

Introduction

There were dramatic price increases that occurred in wholesale electric markets in Midwestern USA during the week of June 22-26, 1998. The price spikes were very high, but short-lived. After questions were raised by a number of market participants, the Federal Energy Regulation Commission (FERC) performed a study to understand why this spike occurred. This case study is based upon FERC’s study.During the week of 25 June, 1998, wholesale electricity prices reached unprecedented levels. The highest price FERC confirmed was an hourly price of US$7 500 per megawatt hour (MWh) for a 50 MW transaction, paid by one Midwest utility for one hour. Prices in the region are usually $25 per MWh.Again in late July 1998, power markets experienced prices that, other than the June 1998 price spike, were the highest of the summer. Many of the factors that contributed to the June spike were also present in July, although the July price increases were not as high and did not last as long as the June spike.

Contributing Factors

It was found that a combination of factors contributed to the June pricing abnormality, including: more demand than supply; defaults on commitments to supply electricity; transmission line loading relief procedures and curtailments; lack of objective price information; market inexperience; lack of demand response by customers; and potential market manipulation.

Findings

Poor Market Information

Many of the buyers apparently did not have good information about what other buyers were paying for hourly power other than by asking sellers or brokers. Without sufficient time to determine the price for power, many utilities appeared to purchase whatever power they could find for whatever price was offered. This may have led them to pay more than they needed to obtain power.


181

Lack of Experience

Market inexperience by participants in commodity markets may have led to some traders to pay higher prices than necessary. Some utilities believed that they could rely on the spot market for the summer rather than making forward commitments on long-term supply deals. Some had sold power forward when supplies seemed ample, but then were caught short during the price spike. Were it not for those commitments, the utilities might have been able to avoid the high prices they paid. Some companies found it difficult to avoid paying high prices during the worst of the crisis because of their lack of experience in dealing in hourly markets. In addition, many market participants did not have creditworthiness checking procedures for enquiring into their counter-parties to ensure contract performance, this subsequently led in a number of instances to defaults occurring on contractual agreements to provide power .General inexperience in competitive markets also appeared when market participants had to deal with their contracts. Awareness of the provisions in their contracts could have saved them a lot of money with regards to liquidated damage provision. Furthermore, disagreements arose about the price of emergency power. There was little understanding of what emergency power is and under what conditions it is sold.

Market Hedging

Several market participants were forced to buy power in hourly markets because their transactions were not properly hedged. If people had sufficiently prepared for the eventuality that people would exercise their options, the defaults might not have taken place. Hedging tools are there to help traders manage the price and counter-party uncertainty. Subsequently, hedging is an essential component of trading in a competitive environment.

Lack of Diversity

Some market participants relied on large packages of power from a single supplier. The loss of one such transaction was enough to force some utilities to try to buy large quantities of power in hourly markets. To ensure delivery of power, market participants should diversify their supplies across a number of regions so that if power cannot move across an interface or if generation outages occur, the participants have other supply alternatives.

Poor Price Signals to Customers

The fact that retail customers had no incentive to adjust their usage based on price contributed to the price spike. Retail Competition — explained in Section 3 of this report — coupled with the ability to respond in real time, could allow customers to see the price of the power they use and react accordingly. Current demand side


182

management measures, such as requests for voluntary cutbacks in electricity use, can help in times of crises, but generally do not provide an incentive for customers to respond effectively to price signals.On Thursday 25 June, the utilities and public officials in Illinois issued public appeals for conservation. These appeals were effective in the sense that voluntary co-operation reduced peak demand when the price spike was at its height. However, if customers had reduced their demand in response to rising prices, the situation might not have become so critical as to require appeals for voluntary action.

Market Manipulation

There were allegations that market participants used various practices to manipulate the wholesale power market in their favour. The practices did not appear to be a direct cause of the price spike, but it diminishes confidence that market institutions are working in a fair and non-discriminatory manner.Ultimately though the adverse circumstances did not compromise the reliability of the electric transmission grid or firm electric service to consumers.

Issues For Consideration

As a result of these events a number of issues /actions have been considered, as detailed below.Centralised trading institutions such as power exchanges and futures markets could have provided better price signals to the market and helped reduce price volatility. Eventually, electricity markets need to move toward electronic trading. Fast-paced hourly power markets require resources for determining prices on an up-to-the minute basis. Accurate and timely information is the key to well-functioning markets. There is a need for co-operation in meeting reliability concerns, and thus consideration should be given to regional co-ordination in areas such as maintenance of transmission and generation systems and transmission planning and operation.Several market participants have publicly indicated that they intend to construct new generation facilities in or around the Midwest. However, this will depend on whether they think they can recover their costs within a reasonable time or not. Thus, the market is resolving the lack of generation capacity, at least over the longer term, on its own without any regulatory action. Both operational and market planning were shown to be important factors in the June event. Following the financial losses incurred during the summer of 1997, many companies developed demand and supply portfolios and other market strategies. Thus, while the market is still immature, market participants are quickly adapting to


183

the new market environment and are undertaking new and innovative risk management measures to minimise risks. There is a requirement to collect the correct data, and the study concluded that future monitoring activities could be improved by having more real time market data readily available to the FERC, as it would facilitate the examination of significant market events. The market itself could benefit as well from more information about electric power transactions. Several market participants asked the FERC to impose general price caps on market-based rates, prices caps on emergency transactions or ‘circuit breakers’ that act as temporary price caps on wholesale prices when they rise. However, FERC believed that price caps, whether applied generally, of intended for specific emergency situations, create a situation in which prices are not allowed to perform their rationing function. The creditworthiness of counter-parties is a major issue brought to light by the Midwest price spike. Apparently, a number of market participants did not have specific or accurate procedures for assessing the creditworthiness of their counter-parties.


If Hong Kong decides to move towards a more competitive electricity market, the incident in the US Midwest during the summer of 1998 offers a number of lessons: market players respond to the incentives provided; transparency and openness are critical to the success of any

competitive market; behaviour cannot be regulated, but market rules can be put in

place; there will be teething problems in any new market; and supply and demand, and not regulators, can determine the

amount of generation that is needed and that will be built.


184

Documents

1 Interconnectio within Hong Kong · Web viewGovernment of the Hong Kong SAR Interconnection and Competition in the Hong Kong Electricity Supply Sector October 1999 Reference C1789